Customized processing for different classes of RDMA traffic

Abstract

A framework for customizing processing for different classes of traffic provisioning is discussed herein. A network device in a communication path between a source host machine and a destination host machine extracts a tag from a packet received by the network device. The packet originates from a source executing on a source host machine and its destination is a destination host machine. The tag is set by the source and indicates a first traffic class to be associated with the packet, the first traffic class being selected by the source from a plurality of traffic classes. The network device determines a first traffic class based on the tag extracted from the packet and processes the packet based on the first traffic class.

Description

[0001] CROSS-REFERENCE TO RELATED APPLICATIONS

[0002] This application claims the benefit of the filing dates of U.S. Provisional Application No. 63 / 332,924, filed on April 20, 2022, and U.S. Non-Provisional Application No. 17 / 935,298, filed on September 26, 2022, each of which is incorporated herein by reference in its entirety for all purposes. Technical Field

[0003] The present disclosure relates to a framework for providing customized processing for different classes of RDMA traffic. In particular, the present disclosure relates to parameterized optimization performed based on different classes of RDMA traffic. Background Art

[0004] Organizations are continually migrating business applications and databases to the cloud to reduce the cost of purchasing, updating, and maintaining on-premises hardware and software. High-performance computing applications are consuming all available computing power to achieve specific outcomes or results. Such applications require dedicated network performance, fast storage, high computing power, and large amounts of memory - resources that are in short supply in the virtualized infrastructure that makes up today's commodity clouds.

[0005] Cloud infrastructure service providers offer newer and faster graphics processing units (GPUs) to address the requirements of these applications. In one case, Remote Direct Memory Access (RDMA) provides high throughput and ultra-low latency with low CPU overhead, which is required for modern data center applications. RDMA is deployed using the RDMA over Converged Ethernet (RoCE) protocol, which relies on priority-based flow control (PFC) to achieve lossless networking. However, RDMA workloads executed on network clusters rarely achieve the desired throughput levels. One of the reasons for this is that the RoCE congestion management protocol executed on the network cluster is not optimized. The embodiments discussed herein address these and other problems. Summary of the invention

[0006] The present disclosure generally relates to a framework for customizing processing for different classes of RDMA traffic. Specifically, the present disclosure relates to parameterized optimization based on different classes of RDMA traffic. RDMA traffic may include: delay-sensitive RDMA traffic, bandwidth-hungry RDMA traffic, etc. Bandwidth-hungry RDMA traffic may also be referred to as bandwidth-sensitive RDMA traffic. As an example, a single queue may be used for delay-sensitive RDMA traffic, while multiple queues may be used for bandwidth-hungry RDMA traffic (i.e., jitter-sensitive traffic) to optimize network traffic. In another example, multiple delay-sensitive traffic classes may be used to avoid scenarios where different types of delay-sensitive traffic cause queues (thus generating jitter) between each other. In other words, multiple delay-sensitive queues may be used to avoid head of line (HOL) blocking scenarios. Various embodiments are described herein, including methods, systems, non-transitory computer-readable storage media storing programs, codes, or instructions executable by one or more processors, etc. These illustrative embodiments are not mentioned to limit or define the present disclosure, but to provide examples to help understand the present disclosure. Additional embodiments are discussed in the detailed description section, and further descriptions are provided therein.

[0007] One embodiment of the present disclosure is directed to a method, comprising: extracting, by a network device in a communication path between a source host machine and a destination host machine, a label from a packet received by the network device, the packet originating from a source executing on the source host machine and destined for the destination host machine, the label being set by the source and indicating a first traffic class to be associated with the packet, the first traffic class being selected by the source from a plurality of traffic classes; determining, by the network device, the first traffic class based on the label extracted from the packet; and processing, by the network device, the packet based on the first traffic class.

[0008] One aspect of the present disclosure provides a network device, which includes one or more data processors and a non-transitory computer-readable storage medium containing instructions, which, when executed on the one or more data processors, cause the one or more data processors to perform part or all of one or more methods disclosed herein.

[0009] Another aspect of the present disclosure provides a computer program product tangibly embodied in a non-transitory machine-readable storage medium, comprising instructions configured to cause one or more data processors to perform part or all of one or more methods disclosed herein.

[0010] The foregoing and other features and embodiments will become more apparent when referring to the following description, claims and accompanying drawings. BRIEF DESCRIPTION OF THE DRAWINGS

[0011] The features, embodiments, and advantages of the present disclosure may be better understood when the following detailed description is read with reference to the accompanying drawings.

[0012] Figure 1 is a high-level diagram of a distributed environment illustrating a virtual or overlay cloud network hosted by a cloud service provider infrastructure according to certain embodiments.

[0013] Figure 2 Depicted is a simplified architectural diagram of physical components in a physical network within a CSPI in accordance with certain embodiments.

[0014] Figure 3 An example arrangement within a CSPI is shown in accordance with certain embodiments, where a host machine is connected to multiple network virtualization devices (NVDs).

[0015] Figure 4 Depicted is connectivity between a host machine and an NVD for providing I / O virtualization to support multitenancy, according to some embodiments.

[0016] Figure 5 Depicted is a simplified block diagram of a physical network provided by CSPI in accordance with certain embodiments.

[0017] Figure 6 An exemplary tree diagram illustrating different parameter settings in accordance with certain embodiments is illustrated.

[0018] Figure 7 An exemplary tree diagram illustrating different setting options for differentiated service code point (DSCP) settings is illustrated in accordance with some embodiments.

[0019] Figure 8 Depicted is a schematic diagram illustrating a mechanism for marking explicit congestion notification (ECN) packets in accordance with certain embodiments.

[0020] Fig. 9 Depicted is a table illustrating different parameter settings for different types of RDMA traffic, in accordance with certain embodiments.

[0021] Fig.10 Depicted is a schematic diagram of identifying network locations in a cloud infrastructure to which different parameter tags are applied, according to certain embodiments.

[0022] Fig.11A An exemplary flow chart depicting steps performed in processing packets based on their traffic class in accordance with certain embodiments is illustrated.

[0023] Fig. 11BAn exemplary flow chart depicting steps performed in processing data packets belonging to a traffic class of a bandwidth sensitive type is illustrated in accordance with certain embodiments.

[0024] Fig. 11C An exemplary flow chart depicting steps performed in processing data packets belonging to a traffic class of a latency-sensitive type is illustrated in accordance with certain embodiments.

[0025] Fig.11D An exemplary flow chart depicting steps performed in tagging portions of a header of a data packet is illustrated in accordance with certain embodiments.

[0026] Fig.12 is a block diagram illustrating one mode for implementing a cloud infrastructure as a service system according to at least one embodiment.

[0027] Fig.13 is a block diagram illustrating another mode for implementing a cloud infrastructure as a service system according to at least one embodiment.

[0028] Fig.14 is a block diagram illustrating another mode for implementing a cloud infrastructure as a service system according to at least one embodiment.

[0029] Fig.15 is a block diagram illustrating another mode for implementing a cloud infrastructure as a service system according to at least one embodiment.

[0030] Fig.16 is a block diagram illustrating an example computer system in accordance with at least one embodiment. DETAILED DESCRIPTION

[0031] In the following description, for the purpose of explanation, specific details are set forth in order to provide a thorough understanding of certain embodiments. However, it is apparent that various embodiments can be practiced without these specific details. The drawings and descriptions are not intended to be limiting. The word "exemplary" is used herein to mean "serving as an example, instance, or illustration." Any embodiment or design described herein as "exemplary" is not necessarily to be construed as preferred or superior to other embodiments or designs.

[0032] Example architecture for cloud infrastructure

[0033] The term cloud service is generally used to refer to services provided on-demand (e.g., via a subscription model) by a cloud service provider (CSP) to users or customers using systems and infrastructure (cloud infrastructure) provided by the CSP. Typically, the servers and systems that make up the CSP's infrastructure are separate from the customer's own on-premises servers and systems. Therefore, customers can leverage the cloud services provided by the CSP themselves without having to purchase separate hardware and software resources for the service. Cloud services are designed to provide subscribing customers with simple, scalable access to applications and computing resources without requiring customers to invest in the infrastructure used to provide the service.

[0034] There are several cloud service providers that offer various types of cloud services. There are various different types or models of cloud services, including software as a service (SaaS), platform as a service (PaaS), infrastructure as a service (IaaS), etc.

[0035] A customer can subscribe to one or more cloud services provided by a CSP. A customer can be any entity, such as an individual, organization, enterprise, etc. When a customer subscribes to or registers for a service provided by a CSP, a lease or account is created for the customer. The customer can then access one or more subscribed cloud resources associated with the account via this account.

[0036] As mentioned above, Infrastructure as a Service (IaaS) is a specific type of cloud computing service. In the IaaS model, the CSP provides infrastructure (called Cloud Service Provider Infrastructure or CSPI) that can be used by customers to build their own customizable networks and deploy customer resources. Therefore, the customer's resources and network are hosted in a distributed environment by the infrastructure provided by the CSP. This is different from traditional computing, in which the customer's resources and network are hosted by the infrastructure provided by the customer.

[0037] CSPI may include interconnected high-performance computing resources, including various host machines, memory resources, and network resources, forming a physical network, which is also referred to as a substrate network or underlying network. The resources in CSPI may be spread across one or more data centers, which may be geographically spread across one or more geographic regions. Virtualization software may be executed by these physical resources to provide a virtualized distributed environment. Virtualization creates an overlay network (also known as a software-based network, a software-defined network, or a virtual network) on a physical network. The CSPI physical network provides an underlying foundation for creating one or more overlay or virtual networks on top of a physical network. A virtual or overlay network may include one or more virtual cloud networks (VCNs). Virtual networks are implemented using software virtualization technology (e.g., hypervisors, functions performed by network virtualization devices (NVDs) (e.g., smartNICs), top-of-rack (TOR) switches, smart TORs that implement one or more functions performed by NVDs, and other mechanisms) to create a layer of network abstraction that can run on top of a physical network. Virtual networks may take a variety of forms, including peer-to-peer networks, IP networks, and the like. Virtual networks are typically layer 3 IP networks or layer 2 VLANs. This approach to virtual or overlay networking is often referred to as virtual or overlay layer 3 networking. Examples of protocols developed for virtual networks include IP-in-IP (or Generic Routing Encapsulation (GRE)), Virtual Extensible LAN (VXLAN—IETF RFC 7348), Virtual Private Network (VPN) (e.g., MPLS Layer 3 Virtual Private Network (RFC 4364)), VMware's NSX, GENEVE (Generic Network Virtualization Encapsulation), and the like.

[0038] For IaaS, the infrastructure (CSPI) provided by the CSP can be configured to provide virtualized computing resources over a public network (e.g., the Internet). In the IaaS model, cloud computing service providers can host infrastructure components (e.g., servers, storage devices, network nodes (e.g., hardware), deployment software, platform virtualization (e.g., hypervisor layer), etc.). In some cases, IaaS providers can also supply various services to accompany these infrastructure components (e.g., billing, monitoring, logging, security, load balancing, and clustering, etc.). Therefore, since these services can be policy-driven, IaaS users can implement policies to drive load balancing to maintain application availability and performance. CSPI provides a collection of infrastructure and complementary cloud services, which enables customers to build and run a wide range of applications and services in a highly available, hosted, distributed environment. CSPI supplies high-performance computing resources and capabilities and storage capacity in a flexible virtual network that can be securely accessed from various networked locations (such as from the customer's local deployment network). When a customer subscribes to or signs up for an IaaS service provided by a CSP, the tenancy created for that customer is a secure and isolated partition within CSPI where the customer can create, organize, and manage their cloud resources.

[0039] Customers can use the computing, storage, and networking resources provided by CSPI to build their own virtual networks. One or more customer resources or workloads, such as computing instances, can be deployed on these virtual networks. For example, customers can use the resources provided by CSPI to build one or more customizable and private virtual networks, called virtual cloud networks (VCNs). Customers can deploy one or more customer resources, such as computing instances, on customer VCNs. Computing instances can take the form of virtual machines, bare metal instances, etc. Therefore, CSPI provides a collection of infrastructure and complementary cloud services that enable customers to build and run a wide range of applications and services in a highly available virtual hosting environment. Customers do not manage or control the underlying physical resources provided by CSPI, but can control operating systems, storage devices, and deployed applications; and may have limited control over selected networking components (e.g., firewalls).

[0040] The CSP may provide a console that enables customers and network administrators to use CSPI resources to configure, access, and manage resources deployed in the cloud. In some embodiments, the console provides a web-based user interface that can be used to access and manage CSPI. In some implementations, the console is a web-based application provided by the CSP.

[0041] CSPI can support single-tenancy or multi-tenancy architecture. In a single-tenancy architecture, software (e.g., application, database) or hardware component (e.g., host machine or server) serves a single customer or tenant. In a multi-tenancy architecture, software or hardware component serves multiple customers or tenants. Therefore, in a multi-tenancy architecture, CSPI resources are shared among multiple customers or tenants. In the case of multi-tenancy, precautions are taken and protections are implemented within CSPI to ensure that each tenant's data is isolated and remains invisible to other tenants.

[0042] In a physical network, a network endpoint ("endpoint") refers to a computing device or system that is connected to a physical network and communicates back and forth with the network to which it is connected. A network endpoint in a physical network can be connected to a local area network (LAN), a wide area network (WAN), or other types of physical networks. Examples of traditional endpoints in a physical network include modems, hubs, bridges, switches, routers and other networking devices, physical computers (or host machines), etc. Each physical device in a physical network has a fixed network address that can be used to communicate with the device. This fixed network address can be a layer 2 address (e.g., a MAC address), a fixed layer 3 address (e.g., an IP address), etc. In a virtualized environment or virtual network, an endpoint can include various virtual endpoints, such as a virtual machine hosted by a component of the physical network (e.g., hosted by a physical host machine). These endpoints in a virtual network are addressed by overlay addresses, such as overlay layer 2 addresses (e.g., overlay MAC addresses) and overlay layer 3 addresses (e.g., overlay IP addresses). Network overlays achieve flexibility by allowing network administrators to move around overlay addresses associated with network endpoints using software management (e.g., via software that implements a control plane for a virtual network). Thus, unlike a physical network, in a virtual network, an overlay address (e.g., an overlay IP address) can be moved from one endpoint to another using network management software. Since the virtual network is built on top of the physical network, the communication between components in the virtual network involves both the virtual network and the underlying physical network. To facilitate such communication, the components of CSPI are configured to learn and store mappings that map overlay addresses in the virtual network to actual physical addresses in the substrate network, and vice versa. These mappings are then used to facilitate communication. Customer services are encapsulated to facilitate routing in the virtual network.

[0043] Thus, a physical address (e.g., a physical IP address) is associated with a component in a physical network, and an overlay address (e.g., an overlay IP address) is associated with an entity in a virtual network. Both a physical IP address and an overlay IP address are types of real IP addresses. These are separate from virtual IP addresses, which are mapped to multiple real IP addresses. A virtual IP address provides a one-to-many mapping between a virtual IP address and multiple real IP addresses.

[0044] The cloud infrastructure or CSPI is physically hosted in one or more data centers in one or more regions of the world. The CSPI may include components in a physical or substrate network and virtualized components (e.g., virtual networks, computing instances, virtual machines, etc.) located in a virtual network built on top of the physical network components. In some embodiments, the CSPI is organized and hosted in realms, regions, and availability domains. A region is typically a localized geographic area containing one or more data centers. Regions are generally independent of each other and can be far apart, for example, across countries or even continents. For example, a first region may be in Australia, another in Japan, another in India, and so on. CSPI resources are divided between regions so that each region has its own independent subset of CSPI resources. Each region can provide a collection of core infrastructure services and resources, such as computing resources (e.g., bare metal servers, virtual machines, containers and related infrastructure, etc.); storage resources (e.g., block volume storage devices, file storage devices, object storage devices, archive storage devices); networking resources (e.g., virtual cloud networks (VCNs), load balancing resources, connections to local deployment networks), database resources; edge networking resources (e.g., DNS); and access management and monitoring resources, etc. Each region generally has multiple paths connecting the region to other regions in the realm.

[0045] Generally speaking, an application is deployed in the region where it is most frequently used (i.e., on the infrastructure associated with that region) because it is faster to use nearby resources than to use distant resources. Applications may also be deployed in different regions for various reasons, such as redundancy to mitigate the risk of region-wide events (such as large weather systems or earthquakes), to meet different requirements of legal jurisdictions, tax domains, and other business or social standards, etc.

[0046] Data centers within a region can be further organized and subdivided into availability domains (ADs). An availability domain can correspond to one or more data centers located within a region. A region can consist of one or more availability domains. In this distributed environment, CSPI resources are either region-specific (such as a virtual cloud network (VCN)) or availability domain-specific (such as a compute instance).

[0047] ADs within a region are isolated from each other, fault-tolerant, and configured so that it is extremely unlikely that they fail at the same time. This is achieved by ADs not sharing key infrastructure resources (such as networking, physical cables, cable paths, cable entry points, etc.), making it unlikely that a failure at one AD within a region will affect the availability of other ADs in the same region. ADs within the same region can be connected to each other through a low-latency, high-bandwidth network, which makes it possible to provide high-availability connections to other networks (e.g., the Internet, the customer's local deployment network, etc.) and to build replication systems in multiple ADs for high availability and disaster recovery. Cloud services use multiple ADs to ensure high availability and prevent resource failures. As the infrastructure provided by the IaaS provider grows, more regions and ADs can be added, as well as additional capacity. Traffic between availability domains is typically encrypted.

[0048] In some embodiments, regions are grouped into domains. A domain is a logical collection of regions. Domains are isolated from each other and do not share any data. Regions in the same domain can communicate with each other, but regions in different domains cannot communicate. A customer's lease or account with a CSP exists in a single domain and can be spread across one or more regions belonging to that domain. Typically, when a customer subscribes to an IaaS service, a lease or account is created for the customer in a customer-specified region within a domain (called the "home" region). The customer can extend the customer's lease to one or more other regions within a domain. A customer cannot access regions that are not in the domain where the customer's lease is located.

[0049] An IaaS provider may offer multiple domains, each catering to a specific set of customers or users. For example, a business domain may be provided for business customers. As another example, a domain may be provided for a specific country for customers within that country. As yet another example, a government domain may be provided for a government, etc. For example, a government domain may cater to a specific government and may have a higher level of security than a business domain. For example, Oracle Cloud Infrastructure (OCI) currently offers a domain for business regions and two domains (e.g., FedRAMP-authorized and IL5-authorized) for government cloud regions.

[0050] In some embodiments, AD can be subdivided into one or more fault domains. Fault domains are groupings of infrastructure resources within an AD to provide anti-affinity. Fault domains allow the distribution of computing instances so that they are not located on the same physical hardware within a single AD. This is called anti-affinity. A fault domain refers to a collection of hardware components (computers, switches, etc.) that share a single point of failure. The computing pool is logically divided into fault domains. Therefore, a hardware failure or computing hardware maintenance event that affects one fault domain does not affect instances in other fault domains. The number of fault domains used for each AD may vary from embodiment to embodiment. For example, in some embodiments, each AD contains three fault domains. Fault domains act as logical data centers within an AD.

[0051] When a customer subscribes to an IaaS service, resources from CSPI are supplied to the customer and associated with the customer's lease. Customers can use these supplied resources to build private networks and deploy resources on these networks. Customer networks hosted by CSPI in the cloud are called virtual cloud networks (VCNs). Customers can use CSPI resources allocated to customers to set up one or more virtual cloud networks (VCNs). VCNs are virtual or software-defined private networks. Customer resources deployed in a customer's VCN may include computing instances (e.g., virtual machines, bare metal instances) and other resources. These computing instances may represent various customer workloads, such as applications, load balancers, databases, and the like. Computing instances deployed on VCNs may communicate with publicly accessible endpoints ("public endpoints"), with other instances in the same VCN or other VCNs (e.g., other VCNs of customers or VCNs that do not belong to the customer), with customer's local deployment data centers or networks, and with service endpoints and other types of endpoints through public networks (such as the Internet).

[0052] CSPs can use CSPI to provide various services. In some cases, customers of CSPI can themselves act like service providers and use CSPI resources to provide services. Service providers can expose service endpoints, which are characterized by identification information (e.g., IP addresses, DNS names, and ports). Customers' resources (e.g., computing instances) can consume a specific service by accessing the service endpoints exposed by the service for a specific service. These service endpoints are generally publicly accessible endpoints to users via public communication networks (such as the Internet) using public IP addresses associated with the endpoints. Publicly accessible network endpoints are sometimes also referred to as public endpoints.

[0053] In some embodiments, a service provider may expose a service via an endpoint for the service (sometimes referred to as a service endpoint). Customers of the service may then use this service endpoint to access the service. In some embodiments, the service endpoint provided for a service may be accessed by multiple customers that intend to consume the service. In other embodiments, a dedicated service endpoint may be provided to a customer so that only the customer can access the service using the dedicated service endpoint.

[0054] In certain embodiments, when a VCN is created, it is associated with a private overlay Classless Inter-Domain Routing (CIDR) address space, which is a range of private overlay IP addresses (e.g., 10.0 / 16) assigned to the VCN. The VCN includes associated subnets, routing tables, and gateways. The VCN resides within a single region but can span one or more or all of the availability domains in that region. A gateway is a virtual interface configured for the VCN and enables communication of traffic to or from the VCN to one or more endpoints outside the VCN. One or more different types of gateways can be configured for the VCN to enable communication to and from different types of endpoints.

[0055] A VCN may be subdivided into one or more subnetworks, such as one or more subnets. Thus, a subnet is a unit or subdivision of a configuration that can be created within a VCN. A VCN may have one or more subnets. Each subnet within a VCN is associated with a contiguous range of overlay IP addresses (e.g., 10.0.0.0 / 24 and 10.0.1.0 / 24) that do not overlap with other subnets in the VCN and represent a subset of the address space within the address space of the VCN.

[0056] Each computing instance is associated with a virtual network interface card (VNIC), which enables the computing instance to participate in the subnet of the VCN. VNIC is a logical representation of a physical network interface card (NIC). In general, a VNIC is an interface between an entity (e.g., a computing instance, a service) and a virtual network. VNICs exist in subnets, have one or more associated IP addresses, and associated security rules or policies. VNICs are equivalent to layer 2 ports on switches. VNICs are attached to subnets within computing instances and VCNs. The VNICs associated with computing instances enable the computing instance to become part of the subnet of the VCN, and enable the computing instance to communicate (e.g., send and receive packets) with endpoints located on the same subnet as the computing instance, with endpoints in different subnets in the VCN, or with endpoints outside the VCN. Therefore, the VNICs associated with computing instances determine how the computing instance is connected to endpoints inside and outside the VCN. When a computing instance is created and added to a subnet within the VCN, a VNIC for the computing instance is created and associated with the computing instance. For a subnet that includes a set of compute instances, the subnet contains VNICs corresponding to the set of compute instances, each VNIC being attached to a compute instance within the set of compute instances.

[0057] A private overlay IP address is assigned to each computing instance via the VNIC associated with the computing instance. This private overlay IP address is assigned to the VNIC associated with the computing instance when the computing instance is created and is used to route the business to and from the computing instance. All VNICs in a given subnet use the same routing table, security list, and DHCP options. As described above, each subnet within a VCN is associated with a continuous range of overlay IP addresses (e.g., 10.0.0.0 / 24 and 10.0.1.0 / 24), which do not overlap with other subnets in the VCN and represent a subset of the address space within the address space of the VCN. For a VNIC on a specific subnet of the VCN, the private overlay IP address assigned to the VNIC is an address from the continuous range of overlay IP addresses allocated for the subnet.

[0058] In some embodiments, in addition to the private overlay IP address, the computing instance may optionally be assigned additional overlay IP addresses, such as one or more public IP addresses in the case of a public subnet, for example. These multiple addresses are assigned on the same VNIC or on multiple VNICs associated with the computing instance. However, each instance has a primary VNIC, which is created during instance startup and is associated with the overlay private IP address assigned to the instance—this primary VNIC cannot be removed. Additional VNICs, called secondary VNICs, can be added to existing instances in the same availability domain as the primary VNIC. All VNICs are in the same availability domain as the instance. The secondary VNIC can be located in a subnet in the same VCN as the primary VNIC, or in a different subnet in the same VCN or in a different VCN.

[0059] If the compute instances are in a public subnet, they can optionally be assigned public IP addresses. When creating a subnet, you can specify the subnet as either a public subnet or a private subnet. A private subnet means that the resources (e.g., compute instances) and associated VNICs in the subnet cannot have public overlay IP addresses. A public subnet means that the resources in the subnet and associated VNICs can have public IP addresses. Customers can specify that a subnet exists in a single availability domain, or across multiple availability domains in a region or domain.

[0060] As described above, a VCN can be subdivided into one or more subnets. In some embodiments, a virtual router (VR) configured for the VCN (referred to as a VCN VR or simply a VR) enables communication between subnets of the VCN. For a subnet within a VCN, the VR represents a logical gateway for the subnet, which enables the subnet (i.e., the computing instances on the subnet) to communicate with endpoints on other subnets within the VCN and other endpoints outside the VCN. A VCN VR is a logical entity that is configured to route traffic between VNICs in the VCN and a virtual gateway ("gateway") associated with the VCN. Figure 1To further describe the gateway. The VCN VR is a layer 3 / IP layer concept. In one embodiment, there is one VCN VR for the VCN, where the VCN VR has a potentially unlimited number of ports addressed by IP addresses, one port for each subnet of the VCN. In this way, the VCN VR has a different IP address for each subnet in the VCN to which the VCN VR is attached. The VR is also connected to various gateways configured for the VCN. In some embodiments, a specific overlay IP address in the overlay IP address range for a subnet is reserved for the ports of the VCN VR for that subnet. For example, consider a VCN with two subnets, with associated address ranges of 10.0 / 16 and 10.1 / 16, respectively. For the first subnet within the VCN with an address range of 10.0 / 16, addresses in this range are reserved for the ports of the VCN VR for that subnet. In some cases, the first IP address in the range can be reserved for the VCN VR. For example, for a subnet with an overlay IP address range of 10.0 / 16, the IP address 10.0.0.1 can be reserved for the ports of the VCN VR for that subnet. For a second subnet within the same VCN with an address range of 10.1 / 16, the VCN VR can have a port for the second subnet with an IP address of 10.1.0.1. The VCN VR has a different IP address for each subnet in the VCN.

[0061] In some other embodiments, each subnet within a VCN may have its own associated VR that may be addressed by the subnet using a reserved or default IP address associated with the VR. For example, the reserved or default IP address may be the first IP address in a range of IP addresses associated with the subnet. The VNICs in the subnet may communicate (e.g., send and receive packets) with the VRs associated with the subnet using this default or reserved IP address. In such embodiments, the VR is the entry / exit point for the subnet. VRs associated with a subnet within the VCN may communicate with other VRs associated with other subnets within the VCN. VRs may also communicate with gateways associated with the VCN. The VR functions for a subnet run on or are performed by one or more NVDs that perform VNIC functions for the VNICs in the subnet.

[0062] Routing tables, security rules, and DHCP options can be configured for a VCN. A routing table is a virtual routing table for a VCN and includes rules for routing traffic from subnets within the VCN to destinations outside the VCN through gateways or specially configured instances. A VCN's routing table can be customized to control how packets are forwarded / routed to and from the VCN. DHCP options refer to configuration information that is automatically provided to an instance when the instance starts.

[0063] The security rules configured for a VCN represent the overlay firewall rules for the VCN. Security rules can include ingress and egress rules and specify the types of services allowed in and out of the VCN instances (e.g., based on protocols and ports). Customers can choose whether a given rule is stateful or stateless. For example, a customer can allow incoming SSH services from anywhere to a collection of instances by setting up a stateful ingress rule with source CIDR 0.0.0.0 / 0 and destination TCP port 22. Security rules can be implemented using network security groups or security lists. A network security group consists of a collection of security rules that apply only to resources in the group. On the other hand, a security list includes rules that apply to all resources in any subnet that uses the security list. A VCN can be set up with a default security list with default security rules. The DHCP options configured for the VCN provide configuration information that is automatically provided to instances in the VCN when the instances are started.

[0064] In some embodiments, configuration information for a VCN is determined and stored by a VCN control plane. For example, configuration information for a VCN may include information about address ranges associated with the VCN, subnets within the VCN and associated information, one or more VRs associated with the VCN, compute instances in the VCN and associated VNICs, NVDs that perform various virtualized network functions associated with the VCN (e.g., VNICs, VRs, gateways), state information for the VCN, and other VCN-related information. In some embodiments, a VCN distribution service publishes configuration information stored by a VCN control plane or a portion thereof to the NVD. The distributed information may be used to update information stored and used by the NVD (e.g., forwarding tables, routing tables, etc.) to forward packets to or from compute instances in the VCN.

[0065] In some embodiments, the creation of VCNs and subnets is handled by the VCN control plane (CP) and the launch of compute instances is handled by the compute control plane. The compute control plane is responsible for allocating physical resources to the compute instances and then calling the VCN control plane to create and attach VNICs to the compute instances. The VCN CP also sends VCN data maps to the VCN data plane, which is configured to perform packet forwarding and routing functions. In some embodiments, the VCN CP provides a distribution service that is responsible for providing updates to the VCN data plane. An example of a VCN control plane is also provided at Fig.12 , Fig.13 , Fig.14 and Fig.15 1216, 1316, 1416, and 1516) and are described below.

[0066] Customers can create one or more VCNs using resources hosted by CSPI. Compute instances deployed on customer VCNs can communicate with different endpoints. These endpoints can include endpoints hosted by CSPI and endpoints external to CSPI.

[0067] Figure 1 , Figure 2 , Figure 3 , Figure 4 , Figure 5 , Fig.12 , Fig.13 , Fig.14 and Fig.15 Various different architectures for implementing cloud-based services using CSPI are depicted in and are described below. Figure 1 is a high-level diagram of a distributed environment 100 showing an overlay or customer VCN hosted by CSPI according to certain embodiments. Figure 1 The distributed environment depicted in includes multiple components in an overlay network. Figure 1 The distributed environment 100 depicted in FIG. 1 is merely an example and is not intended to unduly limit the scope of the claimed embodiments. Many variations, substitutions, and modifications are possible. For example, in some embodiments, Figure 1 The distributed environment depicted in Figure 1 The systems or components shown in FIG. 5 may include more or fewer systems or components, may combine two or more systems, or may have different system configurations or arrangements.

[0068] like Figure 1 As shown in the example depicted in , the distributed environment 100 includes a CSPI 101 that provides services and resources that customers can subscribe to and use to build their virtual cloud networks (VCNs). In some embodiments, the CSPI 101 supplies IaaS services to subscribing customers. The data centers within the CSPI 101 can be organized into one or more regions. Figure 1 An example region "Region US" 102 is shown in FIG. A customer has configured a customer VCN 104 for region 102. The customer can deploy various computing instances on VCN 104, where the computing instances can include virtual machines or bare metal instances. Examples of instances include applications, databases, load balancers, etc.

[0069] exist Figure 1 In the embodiment depicted in , customer VCN 104 includes two subnets, namely, "Subnet-1" and "Subnet-2", each with its own CIDR IP address range. Figure 1, the coverage IP address range of Subnet-1 is 10.0 / 16, and the address range of Subnet-2 is 10.1 / 16. VCN virtual router 105 represents a logical gateway for the VCN that enables communication between subnets of VCN 104 and with other endpoints outside the VCN. VCN VR 105 is configured to route traffic between VNICs in VCN 104 and gateways associated with VCN 104. VCN VR 105 provides ports for each subnet of VCN 104. For example, VR 105 may provide a port with IP address 10.0.0.1 for Subnet-1 and a port with IP address 10.1.0.1 for Subnet-2.

[0070] Multiple computing instances can be deployed on each subnet, where the computing instances can be virtual machine instances and / or bare metal instances. The computing instances in a subnet can be hosted by one or more host machines within CSPI 101. The computing instances participate in the subnet via the VNICs associated with the computing instances. For example, Figure 1 As shown in , compute instance C1 becomes part of subnet-1 via a VNIC associated with the compute instance. Similarly, compute instance C2 becomes part of subnet-1 via a VNIC associated with C2. In a similar manner, multiple compute instances (which may be virtual machine instances or bare metal instances) may be part of subnet-1. Each compute instance is assigned a private overlay IP address and MAC address via its associated VNIC. For example, in Figure 1 , the overlay IP address of compute instance C1 is 10.0.0.2 and the MAC address is M1, while the private overlay IP address of compute instance C2 is 10.0.0.3 and the MAC address is M2. Each compute instance in subnet-1 (including compute instances C1 and C2) has a default route to VCN VR 105 using IP address 10.0.0.1, which is the IP address of the port of VCN VR 105 for subnet-1.

[0071] Multiple computing instances can be deployed on subnet-2, including virtual machine instances and / or bare metal instances. Figure 1 As shown in , compute instances D1 and D2 become part of subnet-2 via the VNICs associated with the respective compute instances. Figure 1In the embodiment depicted in , the overlay IP address of compute instance D1 is 10.1.0.2 and the MAC address is MM1, while the private overlay IP address of compute instance D2 is 10.1.0.3 and the MAC address is MM2. Each compute instance in subnet-2 (including compute instances D1 and D2) has a default route to VCN VR 105 using IP address 10.1.0.1, which is the IP address of the port of VCN VR 105 for subnet-2.

[0072] VCN A 104 may also include one or more load balancers. For example, a load balancer may be provided for a subnet, and the load balancer may be configured to load balance services across multiple computing instances on the subnet. A load balancer may also be provided to load balance services across subnets in the VCN.

[0073] A particular compute instance deployed on VCN 104 can communicate with a variety of different endpoints. These endpoints can include endpoints hosted by CSPI 200 and endpoints external to CSPI 200. Endpoints hosted by CSPI 101 can include: endpoints on the same subnet as a particular compute instance (e.g., communication between two compute instances in subnet-1); endpoints on different subnets but within the same VCN (e.g., communication between a compute instance in subnet-1 and a compute instance in subnet-2); endpoints in different VCNs in the same region (e.g., communication between a compute instance in subnet-1 and an endpoint in VCN 106 or 110 in the same region, communication between a compute instance in subnet-1 and an endpoint in service network 110 in the same region); or endpoints in VCNs in different regions (e.g., communication between a compute instance in subnet-1 and an endpoint in VCN 108 in a different region). Compute instances in a subnet hosted by CSPI 101 may also communicate with endpoints that are not hosted by CSPI 101 (i.e., external to CSPI 101). These external endpoints include endpoints in a customer's on-premises network 116, endpoints within other remote cloud-hosted networks 118, public endpoints 114 accessible via a public network (such as the Internet), and other endpoints.

[0074] Use VNICs associated with source and destination computing instances to facilitate communication between computing instances on the same subnet. For example, computing instance C1 in subnet-1 may want to send a packet to computing instance C2 in subnet-1. For a packet originating from a source computing instance and whose destination is another computing instance in the same subnet, the packet is first processed by the VNIC associated with the source computing instance. The processing performed by the VNIC associated with the source computing instance may include determining the destination information of the packet based on the packet header, identifying any policies (e.g., security lists) configured for the VNIC associated with the source computing instance, determining the next hop for the packet, performing any packet encapsulation / decapsulation functions as needed, and then forwarding / routing the packet to the next hop, with the purpose of facilitating communication of the packet to its intended destination. When the destination computing instance is in the same subnet as the source computing instance, the VNIC associated with the source computing instance is configured to identify the VNIC associated with the destination computing instance and forward the packet to the VNIC for processing. The VNIC associated with the destination computing instance is then executed and the VNIC forwards the packet to the destination computing instance.

[0075] For packets to be transmitted from a compute instance in a subnet to an endpoint in a different subnet in the same VCN, communication is facilitated through the VNICs associated with the source and destination compute instances and the VCN VRs. Figure 1 If compute instance C1 in subnet-1 wants to send a packet to compute instance D1 in subnet-2, the packet is first processed by the VNIC associated with compute instance C1. The VNIC associated with compute instance C1 is configured to route the packet to VCN VR 105 using the default route or port 10.0.0.1 of the VCN VR. VCN VR 105 is configured to route the packet to subnet-2 using port 10.1.0.1. The VNIC associated with D1 then receives and processes the packet and the VNIC forwards the packet to compute instance D1.

[0076] For packets to be transmitted from a compute instance in VCN 104 to an endpoint outside of VCN 104, communication is facilitated by a VNIC associated with the source compute instance, a VCN VR 105, and a gateway associated with VCN 104. One or more types of gateways may be associated with VCN 104. A gateway is an interface between a VCN and another endpoint, where the other endpoint is outside of the VCN. A gateway is a layer 3 / IP layer concept and enables a VCN to communicate with an endpoint outside of the VCN. Thus, a gateway facilitates the flow of traffic between a VCN and other VCNs or networks. A variety of different types of gateways may be configured for a VCN to facilitate different types of communications with different types of endpoints. Depending on the gateway, communications may be conducted over a public network (e.g., the Internet) or over a private network. Various communication protocols may be used for these communications.

[0077] For example, compute instance C1 may want to communicate with an endpoint outside of VCN 104. The packet may first be processed by a VNIC associated with source compute instance C1. The VNIC processing determines that the destination of the packet is outside of C1's subnet-1. The VNIC associated with C1 may forward the packet to VCN VR 105 for VCN 104. VCN VR 105 then processes the packet and, as part of the processing, determines a specific gateway associated with VCN 104 as the next hop for the packet based on the packet's destination. VCN VR 105 may then forward the packet to the specific identified gateway. For example, if the destination is an endpoint within a customer's on-premises network, the packet may be forwarded by VCN VR 105 to a dynamic routing gateway (DRG) gateway 122 configured for VCN 104. The packet may then be forwarded from the gateway to the next hop to facilitate delivery of the packet to its final intended destination.

[0078] A variety of different types of gateways can be configured for a VCN. Examples of gateways that can be configured for a VCN are in Figure 1 An example of a gateway associated with a VCN is also shown in FIG. Fig.12 , Fig.13 , Fig.14 and Fig.15 , (e.g., gateways referenced by reference numerals 1234, 1236, 1238, 1334, 1336, 1338, 1434, 1436, 1438, 1534, 1536, and 1538) and are described below. Figure 1As shown in the embodiment depicted in , a dynamic routing gateway (DRG) 122 can be added to or associated with a customer VCN 104 and provide a path for private network traffic communication between the customer VCN 104 and another endpoint, where the other endpoint can be a customer's on-premises network 116, a VCN 108 in a different region of CSPI 101, or other remote cloud network 118 not hosted by CSPI 101. The customer's on-premises network 116 can be a customer network or a customer data center built using the customer's resources. Access to the customer's on-premises network 116 is generally very limited. For customers who have both a customer's on-premises network 116 and one or more VCNs 104 deployed or hosted in the cloud by CSPI 101, the customer may want their on-premises network 116 and their cloud-based VCN 104 to be able to communicate with each other. This enables customers to build an extended hybrid environment that includes the customer's VCN 104 hosted by CSPI 101 and their on-premises network 116. DRG 122 enables such communication. To achieve such communication, a communication channel 124 is set up, where one endpoint of the channel is located in the customer's on-premises network 116, and the other endpoint is located in the CSPI 101 and connected to the customer VCN 104. The communication channel 124 can pass through a public communication network (such as the Internet) or a private communication network. A variety of different communication protocols can be used, such as IPsec VPN technology over a public communication network (such as the Internet), Oracle's FastConnect technology that uses a private network instead of a public network, etc. The equipment or equipment that forms one endpoint of the communication channel 124 in the customer's on-premises network 116 is called customer premises equipment (CPE), such as Figure 1 On the CSPI 101 side, the endpoint may be a host machine executing DRG 122.

[0079] In some embodiments, a remote peer connection (RPC) can be added to the DRG, which allows a customer to peer one VCN with another VCN in a different region. Using this RPC, a customer VCN 104 can connect to a VCN 108 in another region using a DRG 122. The DRG 122 can also be used to communicate with other remote cloud networks 118 that are not hosted by the CSPI 101, such as Microsoft Azure Cloud, Amazon AWS Cloud, etc.

[0080] like Figure 1As shown in , an Internet Gateway (IGW) 120 can be configured for a customer VCN 104, which enables computing instances on the VCN 104 to communicate with public endpoints 114 accessible through a public network, such as the Internet. The IGW 1120 is a gateway that connects the VCN to a public network, such as the Internet. The IGW 120 enables public subnets within a VCN, such as VCN 104, where resources in the public subnet have public overlay IP addresses, to directly access public endpoints 112 on a public network 114, such as the Internet. Using the IGW 120, a connection can be initiated from a subnet within the VCN 104 or from the Internet.

[0081] A network address translation (NAT) gateway 128 may be configured for a customer's VCN 104 and enables cloud resources in the customer's VCN that do not have dedicated public overlay IP addresses to access the Internet, and do so without exposing those resources to direct incoming Internet connections (e.g., L4-L7 connections). This enables private subnets within a VCN (such as private subnet-1 in VCN 104) to privately access public endpoints on the Internet. In a NAT gateway, connections can only be initiated from a private subnet to the public Internet, and connections cannot be initiated from the Internet to a private subnet.

[0082] In some embodiments, a service gateway (SGW) 126 may be configured for a customer VCN 104 and provide a path for private network traffic between the VCN 104 and service endpoints supported in the service network 110. In some embodiments, the service network 110 may be provided by a CSP and may provide a variety of services. An example of such a service network is Oracle's service network, which provides a variety of services available to customers. For example, a computing instance (e.g., a database system) in a private subnet of a customer VCN 104 may back up data to a service endpoint (e.g., an object storage device) without requiring a public IP address or access to the Internet. In some embodiments, a VCN may have only one SGW, and connections may only be initiated from a subnet within the VCN, not from the service network 110. If a VCN is peered with another VCN, resources in the other VCN generally cannot access the SGW. Resources in a local deployment network connected to the VCN using FastConnect or VPN Connect may also use the service gateway configured for the VCN.

[0083] In some embodiments, the SGW 126 uses the concept of a service classless inter-domain routing (CIDR) label, which is a string that represents all regional public IP address ranges for a service or group of services of interest. Customers use the service CIDR label when they configure the SGW and associated routing rules to control traffic to the service. If the public IP addresses of the services change in the future, customers can optionally use the service CIDR label when configuring security rules without having to adjust the security rules.

[0084] A Local Peer Gateway (LPG) 132 is a gateway that can be added to a customer VCN 104 and enables the VCN 104 to peer with another VCN in the same region. Peering refers to the VCNs communicating using private IP addresses without the traffic having to traverse a public network such as the Internet or without routing the traffic through the customer's on-premises network 116. In a preferred embodiment, the VCN has a separate LPG for each peer it establishes. Local peering or VCN peering is a common practice for establishing network connectivity between different applications or infrastructure management functions.

[0085] Service providers (such as providers of services in service network 110) can provide access to services using different access models. According to the public access model, the service can be exposed as a public endpoint that can be publicly accessed by a computing instance in a customer VCN via a public network (such as the Internet), and / or can be privately accessed via SGW 126. According to a specific private access model, the service can be accessed as a private IP endpoint in a private subnet in the customer's VCN. This is called private endpoint (PE) access and enables service providers to expose their services as instances in the customer's private network. Private endpoint resources represent services within the customer's VCN. Each PE appears as a VNIC (called a PE-VNIC, with one or more private IPs) in a subnet selected by the customer in the customer's VCN. Therefore, PE provides a way to present services in a private customer VCN subnet using VNICs. Since the endpoint is exposed as a VNIC, all features associated with the VNIC (such as routing rules, security lists, etc.) are now available to the PE VNIC.

[0086] Service providers can register their services to enable access through PEs. Providers can associate policies with services, which limits the visibility of the service to customer tenancies. Providers can register multiple services under a single virtual IP address (VIP), especially for multi-tenant services. There can be multiple such private endpoints (in multiple VCNs) representing the same service.

[0087] The computing instances in the private subnet can then access the service using the private IP address or service DNS name of the PE VNIC. The computing instances in the customer VCN can access the service by sending the business to the private IP address of the PE in the customer VCN. The private access gateway (PAGW) 130 is a gateway resource that can be attached to a service provider VCN (e.g., a VCN in the service network 110) that acts as an entry / exit point for all businesses from / to the private endpoints of the customer subnet. PAGW 130 enables providers to expand the number of PE connections without utilizing their internal IP address resources. Providers only need to configure one PAGW for any number of services registered in a single VCN. Providers can represent services as private endpoints in multiple VCNs for one or more customers. From the customer's perspective, instead of being attached to the customer's instance, the PE VNIC appears to be attached to the service that the customer wants to interact with. Business to the private endpoint is routed to the service via PAGW 130. These are called customer-to-service private connections (C2S connections).

[0088] The PE concept can also be used to extend private access to services to the customer's on-premises networks and data centers by allowing traffic to flow through FastConnect / IPsec links and private endpoints in the customer's VCN. Private access to services can also be extended to the customer's peer VCNs by allowing traffic to flow between the LPG 132 and the PEs in the customer's VCN.

[0089] The customer can control routing in the VCN at the subnet level, so the customer can specify which subnets in the customer's VCN (such as VCN 104) use each gateway. The routing table of the VCN is used to decide whether to allow traffic to leave the VCN through a particular gateway. For example, in a specific example, the routing table for a public subnet within a customer VCN 104 can send non-local traffic through IGW 120. The routing table for a private subnet within the same customer VCN 104 can send traffic destined for CSP services through SGW 126. All remaining traffic can be sent via NAT gateway 128. The routing table only controls traffic leaving the VCN.

[0090] The security list associated with the VCN is used to control the business that enters the VCN via the gateway via an inbound connection. All resources in the subnet use the same routing table and security list. The security list can be used to control the specific types of business that are allowed to enter and exit the instance in the subnet of the VCN. The security list rules may include ingress (inbound) and egress (outbound) rules. For example, the ingress rule may specify the allowed source address range, and the egress rule may specify the allowed destination address range. The security rules may specify specific protocols (e.g., TCP, ICMP), specific ports (e.g., 22 for SSH, 3389 for Windows RDP), etc. In some embodiments, the operating system of the instance may enforce its own firewall rules that comply with the security list rules. The rules may be stateful (e.g., tracking connections and automatically allowing responses without explicit security list rules for the response business) or stateless.

[0091] Access from a customer VCN (i.e., through resources or compute instances deployed on VCN 104) can be categorized as public access, private access, or dedicated access. Public access refers to an access model that uses a public IP address or NAT to access a public endpoint. Private access enables customer workloads in VCN 104 (e.g., resources in a private subnet) with private IP addresses to access services without traversing a public network such as the Internet. In certain embodiments, CSPI 101 enables customer VCN workloads with private IP addresses to access (public service endpoints of) services using a service gateway. Thus, the service gateway provides a private access model by establishing a virtual link between a customer's VCN and a public endpoint of a service that resides outside of the customer's private network.

[0092] In addition, CSPI can use technologies such as FastConnect public peering to provision dedicated public access, where customer on-premises instances can use FastConnect connections to access one or more services in a customer VCN without traversing a public network (such as the Internet). CSPI can also use FastConnect private peering to provision dedicated private access, where customer on-premises instances with private IP addresses can use FastConnect connections to access customer VCN workloads. FastConnect is a network connectivity alternative to using the public Internet to connect a customer's on-premises network to CSPI and its services. FastConnect provides a simple, flexible, and cost-effective way to create dedicated and private connections with higher bandwidth options and a more reliable and consistent networking experience than Internet-based connections.

[0093] Figure 1and the accompanying description above describe various virtualized components in an example virtual network.As described above, a virtual network is built on an underlying physical or substrate network. Figure 2 A simplified architecture diagram of physical components in a physical network within a CSPI 200 that provides an underlying virtual network according to certain embodiments is depicted. As shown, the CSPI 200 provides a distributed environment that includes components and resources (e.g., computing, storage, and networking resources) provided by a cloud service provider (CSP). These components and resources are used to provide cloud services (e.g., IaaS services) to subscribing customers (i.e., customers who have subscribed to one or more services provided by the CSP). Based on the services subscribed by the customer, a subset of the resources (e.g., computing, storage, and networking resources) of the CSPI 200 is provisioned to the customer. The customer can then build its own cloud-based (i.e., CSPI-hosted) customizable and private virtual network using the physical computing, storage, and networking resources provided by the CSPI 200. As indicated previously, these customer networks are referred to as virtual cloud networks (VCNs). Customers can deploy one or more customer resources, such as computing instances, on these customer VCNs. The computing instances can be in the form of virtual machines, bare metal instances, etc. The CSPI 200 provides a collection of infrastructure and complementary cloud services that enable customers to build and run a wide range of applications and services in a highly available hosting environment.

[0094] exist Figure 2 In the example embodiment depicted in , the physical components of CSPI 200 include one or more physical host machines or physical servers (e.g., 202, 206, 208), network virtualization devices (NVDs) (e.g., 210, 212), top-of-rack (TOR) switches (e.g., 214, 216), and physical networks (e.g., 218), and switches in physical network 218. The physical host machines or servers can host and execute various computing instances that participate in one or more subnets of the VCN. The computing instances can include virtual machine instances and bare metal instances. For example, Figure 1 The various computational examples depicted in Figure 2 The physical host machines depicted in the figure host the virtual machine compute instances in the VCN. The virtual machine compute instances in the VCN can be executed by one host machine or multiple different host machines. The physical host machines can also host virtual host machines, container-based hosts or functions, etc. Figure 1 The VNIC and VCN VR depicted in the figure can be represented by Figure 2 The NVD implementation depicted in FIG. Figure 1 The gateway depicted in can be Figure 2 The host machine depicted in and / or executed by NVD.

[0095] A host machine or server may execute a hypervisor (also referred to as a virtual machine monitor or VMM) that creates and enables a virtualized environment on the host machine. Virtualization or a virtualized environment facilitates cloud-based computing. One or more computing instances may be created, executed, and managed on the host machine by a hypervisor on the host machine. The hypervisor on the host machine enables the physical computing resources (e.g., computing, memory, and networking resources) of the host machine to be shared between the various computing instances executed by the host machine.

[0096] For example, Figure 2 As depicted in , host machines 202 and 208 execute hypervisors 260 and 266, respectively. These hypervisors may be implemented using software, firmware, or hardware, or a combination thereof. Typically, a hypervisor is a process or software layer that sits on top of an operating system (OS) of a host machine, which in turn executes on a hardware processor of the host machine. A hypervisor provides a virtualized environment by enabling the physical computing resources of a host machine (e.g., processing resources such as processors / cores, memory resources, networking resources) to be shared among various virtual machine computing instances executed by the host machine. For example, in Figure 2 In the present invention, the hypervisor 260 can be located on top of the OS of the host machine 202 and enable the computing resources (e.g., processing, memory, and networking resources) of the host machine 202 to be shared between computing instances (e.g., virtual machines) executed by the host machine 202. The virtual machine can have its own operating system (called a guest operating system), which can be the same as or different from the OS of the host machine. The operating system of a virtual machine executed by the host machine can be the same as or different from the operating system of another virtual machine executed by the same host machine. Thus, the hypervisor enables multiple operating systems to execute alongside each other while sharing the same computing resources of the host machine. Figure 2 The host machines depicted in FIG. 1 may have the same or different types of hypervisors.

[0097] Compute instances can be virtual machine instances or bare metal instances. Figure 2 , compute instance 268 on host machine 202 and compute instance 274 on host machine 208 are examples of virtual machine instances. Host machine 206 is an example of a bare metal instance provided to a customer.

[0098] In some cases, the entire host machine can be supplied to a single customer, and all one or more computing instances (or virtual machines or bare metal instances) hosted by the host machine belong to the same customer. In other cases, the host machine can be shared between multiple customers (i.e., multiple tenants). In this multi-tenant scenario, the host machine can host virtual machine computing instances belonging to different customers. These computing instances can be members of different VCNs of different customers. In certain embodiments, the bare metal computing instance is hosted by a bare metal server without a hypervisor. When supplying a bare metal computing instance, a single customer or tenant maintains control of the physical CPU, memory, and network interface of the host machine hosting the bare metal instance, and the host machine is not shared with other customers or tenants.

[0099] As previously described, each compute instance that is part of a VCN is associated with a VNIC that enables the compute instance to be a member of a subnet of the VCN. The VNIC associated with a compute instance facilitates the communication of packets or frames to and from the compute instance. The VNIC is associated with the compute instance when the compute instance is created. In some embodiments, for a compute instance executed by a host machine, the VNIC associated with the compute instance is executed by an NVD connected to the host machine. For example, in Figure 2 , host machine 202 executes virtual machine compute instance 268 associated with VNIC 276, and VNIC 276 is executed by NVD 210 connected to host machine 202. As another example, bare metal instance 272 hosted by host machine 206 is associated with VNIC 280 executed by NVD 212 connected to host machine 206. As yet another example, VNIC 284 is associated with compute instance 274 executed by host machine 208, and VNIC 284 is executed by NVD 212 connected to host machine 208.

[0100] For compute instances hosted by a host machine, the NVD connected to the host machine also executes the VCN VR corresponding to the VCN of which the compute instance is a member. Figure 2 In the embodiment depicted in , NVD 210 executes VCN VR 277 corresponding to the VCN of which compute instance 268 is a member. NVD 212 may also execute one or more VCN VRs 283 corresponding to the VCNs corresponding to the compute instances hosted by host machines 206 and 208.

[0101] The host machine may include one or more network interface cards (NICs) that enable the host machine to connect to other devices. The NIC on the host machine may provide one or more ports (or interfaces) that enable the host machine to communicatively connect to another device. For example, the host machine may connect to the NVD using one or more ports (or interfaces) provided on the host machine and on the NVD. The host machine may also connect to other devices (such as another host machine).

[0102] For example, in Figure 2 2, host machine 202 is connected to NVD 210 using link 220, which extends between port 234 provided by NIC 232 of host machine 202 and port 236 of NVD 210. Host machine 206 is connected to NVD 212 using link 224, which extends between port 246 provided by NIC 244 of host machine 206 and port 248 of NVD 212. Host machine 208 is connected to NVD 212 using link 226, which extends between port 252 provided by NIC 250 of host machine 208 and port 254 of NVD 212.

[0103] The NVD is in turn connected via communication links to top-of-rack (TOR) switches, which are connected to the physical network 218 (also referred to as a switch fabric). In some embodiments, the links between the host machine and the NVD and between the NVD and the TOR switches are Ethernet links. For example, in Figure 2 , NVDs 210 and 212 are connected to TOR switches 214 and 216 using links 228 and 230, respectively. In some embodiments, links 220, 224, 226, 228, and 230 are Ethernet links. A collection of host machines and NVDs connected to a TOR is sometimes referred to as a rack.

[0104] Physical network 218 provides a communication architecture that enables TOR switches to communicate with each other. Physical network 218 can be a multi-layer network. In some embodiments, physical network 218 is a multi-layer Clos network of switches, where TOR switches 214 and 216 represent leaf nodes of the multi-layer and multi-node physical switching network 218. Different Clos network configurations are possible, including but not limited to 2-layer networks, 3-layer networks, 4-layer networks, 5-layer networks, and general "n"-layer networks. Examples of Clos networks are described in Figure 5 and described below.

[0105] There may be various connection configurations between the host machine and the NVD, such as one-to-one configuration, many-to-one configuration, one-to-many configuration, etc. In a one-to-one configuration implementation, each host machine is connected to its own separate NVD. Figure 2 In , host machine 202 is connected to NVD 210 via NIC 232 of host machine 202. In a many-to-one configuration, multiple host machines are connected to one NVD. Figure 2 , host machines 206 and 208 are connected to the same NVD 212 via NICs 244 and 250, respectively.

[0106] In a one-to-many configuration, one host machine connects to multiple NVDs. Figure 3 An example within CSPI 300 is shown, where a host machine is connected to multiple NVDs. Figure 3 , a host machine 302 includes a network interface card (NIC) 304 that includes a plurality of ports 306 and 308. The host machine 300 is connected to a first NVD 310 via port 306 and link 320, and to a second NVD 312 via port 308 and link 322. Ports 306 and 308 may be Ethernet ports and links 320 and 322 between the host machine 302 and the NVDs 310 and 312 may be Ethernet links. The NVD 310 is in turn connected to a first TOR switch 314 and the NVD 312 is connected to a second TOR switch 316. The links between the NVDs 310 and 312 and between the TOR switches 314 and 316 may be Ethernet links. The TOR switches 314 and 316 represent layer 0 switching devices in a multi-layer physical network 318.

[0107] Figure 3 The arrangement depicted in FIG. 3 provides two separate physical network paths between the physical switch network 318 and the host machine 302: the first path passes through the TOR switch 314 to the NVD 310 and then to the host machine 302, and the second path passes through the TOR switch 316 to the NVD 312 and then to the host machine 302. The separate paths provide enhanced availability (referred to as high availability) of the host machine 302. If there is a problem with one of the following: the path (e.g., a link in one of the paths is broken) or the device (e.g., a particular NVD is not running), then the other path can be used for communication to / from the host machine 302.

[0108] exist Figure 3 In the configuration depicted in , the host machine is connected to two different NVDs using two different ports provided by the host machine's NIC. In other embodiments, the host machine may include multiple NICs that enable the host machine to connect to multiple NVDs.

[0109] Return to reference Figure 2NVD is a physical device or component that performs one or more network and / or storage virtualization functions. NVD can be any device with one or more processing units (e.g., CPU, network processing unit (NPU), FPGA, packet processing pipeline, etc.), memory (including cache), and ports. Various virtualization functions can be performed by software / firmware executed by one or more processing units of NVD.

[0110] NVD can be implemented in a variety of different forms. For example, in some embodiments, NVD is implemented as an interface card called a smartNIC or an intelligent NIC with an onboard embedded processor. A smartNIC is a device that is independent of the NIC on the host machine. Figure 2 , NVDs 210 and 212 may be implemented as smartNICs connected to host machine 202 and host machines 206 and 208, respectively.

[0111] However, smartNIC is only one example of NVD implementation. Various other implementations are possible. For example, in some other implementations, NVD or one or more functions performed by NVD can be incorporated into or performed by one or more host machines, one or more TOR switches, and other components of CSPI 200. For example, NVD can be implemented in a host machine, where the functions performed by NVD are performed by the host machine. As another example, NVD can be part of a TOR switch, or a TOR switch can be configured to perform the functions performed by NVD, which enables the TOR switch to perform various complex packet conversions for public clouds. TORs that perform the functions of NVD are sometimes referred to as smart TORs. In other implementations where virtual machine (VM) instances are supplied to customers instead of bare metal (BM) instances, the functions performed by NVD can be implemented inside the hypervisor of the host machine. In some other implementations, some of the functions of NVD can be offloaded to centralized services running on a group of host machines.

[0112] In some embodiments, such as when implemented as Figure 2 When a smartNIC is shown in FIG. 1 , the NVD may include multiple physical ports that enable the NVD to connect to one or more host machines and one or more TOR switches. Ports on the NVD may be classified as host-facing ports (also referred to as "south ports"), or network-facing or TOR-facing ports (also referred to as "north ports"). The host-facing ports of the NVD are the ports used to connect the NVD to a host machine. Figure 2Examples of host-facing ports include port 236 on NVD 210 and ports 248 and 254 on NVD 212. The network-facing ports of the NVD are the ports used to connect the NVD to the TOR switch. Figure 2 Examples of network-facing ports include port 256 on NVD 210 and port 258 on NVD 212. Figure 2 As shown in FIG. 1 , NVD 210 is connected to TOR switch 214 using link 228 extending from port 256 of NVD 210 to TOR switch 214. Similarly, NVD 212 is connected to TOR switch 216 using link 230 extending from port 258 of NVD 212 to TOR switch 216.

[0113] NVD receives packets and frames (e.g., packets and frames generated by a compute instance hosted by the host machine) from a host machine via a host-facing port, and after performing necessary packet processing, may forward the packets and frames to a TOR switch via a network-facing port of NVD. NVD may receive packets and frames from a TOR switch via a network-facing port of NVD, and after performing necessary packet processing, may forward the packets and frames to a host machine via a host-facing port of NVD.

[0114] In some embodiments, there may be multiple ports and associated links between the NVD and the TOR switch. These ports and links may be aggregated to form a link aggregate group (referred to as a LAG) of multiple ports or links. Link aggregation allows multiple physical links between two endpoints (e.g., between the NVD and the TOR switch) to be treated as a single logical link. All physical links in a given LAG may operate in full-duplex mode at the same speed. LAG helps increase the bandwidth and reliability of the connection between the two endpoints. If one of the physical links in the LAG fails, the service will be dynamically and transparently reassigned to one of the other physical links in the LAG. The aggregated physical links deliver higher bandwidth than each individual link. Multiple ports associated with the LAG are treated as a single logical port. Services may be load balanced between multiple physical links in the LAG. One or more LAGs may be configured between two endpoints. The two endpoints may be located between the NVD and the TOR switch, between the host machine and the NVD, and so on.

[0115] NVD implements or performs network virtualization functions. These functions are performed by software / firmware executed by NVD. Examples of network virtualization functions include, but are not limited to: packet encapsulation and decapsulation functions; functions for creating VCN networks; functions for implementing network policies, such as VCN security lists (firewall) functions; functions for facilitating routing and forwarding of packets to and from computing instances in the VCN, etc. In some embodiments, after receiving a packet, NVD is configured to execute a packet processing pipeline for processing the packet and determining how to forward or route the packet. As part of this packet processing pipeline, NVD can execute one or more virtual functions associated with the overlay network, such as executing a VNIC associated with a cis in the VCN, executing a virtual router (VR) associated with the VCN, encapsulation and decapsulation of packets to facilitate forwarding or routing in the virtual network, execution of certain gateways (e.g., local peer gateways), implementation of security lists, network security groups, network address translation (NAT) functions (e.g., translating public IPs to private IPs on a host-by-host basis), throttling functions, and other functions.

[0116] In some embodiments, the packet processing data path in NVD may include multiple packet pipelines, each packet pipeline consisting of a series of packet transformation stages. In some embodiments, after receiving a packet, the packet is parsed and classified into a single pipeline. The packet is then processed in a linear manner, one stage after another, until the packet is discarded or sent out through the interface of NVD. These stages provide basic functional packet processing building blocks (e.g., verifying headers, enforcing throttling, inserting new layer 2 headers, enforcing L4 firewalls, VCN encapsulation / decapsulation, etc.) so that new pipelines can be built by combining existing stages, and new functionality can be added by creating new stages and plugging them into existing pipelines.

[0117] NVD can perform both control plane functions and data plane functions corresponding to the control plane and data plane of the VCN. An example of a VCN control plane is also provided in Fig.12 , Fig.13 , Fig.14 and Fig.15 An example of a VCN data plane is depicted in FIG. 1 (see reference numerals 1216, 1316, 1416, and 1516) and described below. Fig.12 , Fig.13 , Fig.14 and Fig.151218, 1318, 1418, and 1518) and described below. Control plane functions include functions for configuring the network that controls how data is forwarded (e.g., setting routes and routing tables, configuring VNICs, etc.). In some embodiments, a VCN control plane is provided that centrally calculates all overlay-to-substrate mappings and publishes them to NVD and virtual network edge devices (such as various gateways, such as DRG, SGW, IGW, etc.). Firewall rules can also be published using the same mechanism. In some embodiments, NVD only obtains mappings associated with that NVD. Data plane functions include functions for actually routing / forwarding packets based on the configuration set using the control plane. The VCN data plane is implemented by encapsulating the customer's network packets before they traverse the substrate network. The encapsulation / decapsulation functions are implemented on the NVD. In some embodiments, the NVD is configured to intercept all network packets entering and leaving the host machine and perform network virtualization functions.

[0118] As indicated above, NVD performs various virtualization functions, including VNIC and VCN VR. NVD can execute VNICs associated with computing instances hosted by one or more host machines connected to the VNIC. For example, Figure 2 , NVD 210 executes the functions of VNIC 276 associated with compute instance 268 hosted by host machine 202 connected to NVD 210. As another example, NVD 212 executes VNIC 280 associated with bare metal compute instance 272 hosted by host machine 206, and executes VNIC 284 associated with compute instance 274 hosted by host machine 208. The host machines may host compute instances belonging to different VCNs (belonging to different customers), and the NVD connected to the host machines may execute the VNICs corresponding to the compute instances (i.e., perform VNIC-related functions).

[0119] NVD also implements a VCN virtual router corresponding to the VCN of the compute instance. Figure 2 , NVD 210 executes VCN VR 277 corresponding to the VCN to which compute instance 268 belongs. NVD 212 executes one or more VCN VRs 283 corresponding to one or more VCNs to which compute instances hosted by host machines 206 and 208 belong. In some embodiments, the VCN VR corresponding to the VCN is executed by all NVDs connected to the host machine hosting at least one compute instance belonging to the VCN. If the host machine hosts compute instances belonging to different VCNs, then the NVDs connected to the host machine can execute VCN VRs corresponding to these different VCNs.

[0120] In addition to the VNICs and VCN VRs, NVD may also execute various software (e.g., daemons) and include one or more hardware components that facilitate the various network virtualization functions performed by NVD. For simplicity, these various components are grouped together as Figure 2 . For example, NVD 210 includes a packet processing component 286 and NVD 212 includes a packet processing component 288. For example, a packet processing component for NVD may include a packet processor configured to interact with the port and hardware interface of NVD to monitor all packets received by NVD and transmitted using NVD and store network information. Network information may, for example, include network flow information and per-flow information (e.g., per-flow statistics) that identify different network flows handled by NVD. In some embodiments, network flow information may be stored on a per-VNIC basis. The packet processor may perform per-packet manipulation and implement stateful NAT and L4 firewall (FW). As another example, the packet processing component may include a replication agent configured to copy information stored by NVD to one or more different replication target repositories. As yet another example, the packet processing component may include a logging agent configured to perform the logging function of NVD. The packet processing component may also include software for monitoring the performance and health of NVD and possibly also monitoring the status and health of other components connected to NVD.

[0121] Figure 1 Components of an example virtual or overlay network are shown, including a VCN, subnets within the VCN, compute instances deployed on the subnets, VNICs associated with the compute instances, VRs for the VCN, and a set of gateways configured for the VCN. Figure 1 The overlay component depicted in can be represented by Figure 2 The compute instances in a VCN can be executed or hosted by one or more of the physical components depicted in FIG. Figure 2 . For a compute instance hosted by a host machine, a VNIC associated with the compute instance is typically executed by an NVD connected to the host machine (i.e., the VNIC functionality is provided by the NVD connected to the host machine). VCN VR functionality for a VCN is executed by all NVDs connected to the host machines that host or execute compute instances as part of the VCN. Gateways associated with a VCN may be executed by one or more different types of NVDs. For example, some gateways may be executed by smartNICs, while other gateways may be executed by one or more host machines or other implementations of NVDs.

[0122] As described above, a compute instance in a customer VCN can communicate with a variety of different endpoints, where the endpoint can be in the same subnet as the source compute instance, in a different subnet than the source compute instance but in the same VCN, or with an endpoint located outside the VCN of the source compute instance. These communications are facilitated using VNICs associated with the compute instances, VCN VRs, and gateways associated with the VCN.

[0123] For communication between two computing instances on the same subnet in a VCN, VNICs associated with the source and destination computing instances are used to facilitate communication. The source and destination computing instances can be hosted by the same host machine or different host machines. Packets originating from the source computing instance can be forwarded from the host machine hosting the source computing instance to the NVD connected to the host machine. On the NVD, packets are processed using a packet processing pipeline, which can include the execution of a VNIC associated with the source computing instance. Since the destination endpoint for the packet is located in the same subnet, the execution of the VNIC associated with the source computing instance causes the packet to be forwarded to the NVD that executes the VNIC associated with the destination computing instance, and then the NVD processes the packet and forwards it to the destination computing instance. The VNICs associated with the source and destination computing instances can be executed on the same NVD (e.g., when both the source and destination computing instances are hosted by the same host machine) or on different NVDs (e.g., when the source and destination computing instances are hosted by different host machines connected to different NVDs). The VNIC can use the routing / forwarding table stored by the NVD to determine the next hop of the packet.

[0124] For packets to be transmitted from a compute instance in a subnet to an endpoint in a different subnet in the same VCN, packets originating from the source compute instance are transmitted from the host machine hosting the source compute instance to the NVD connected to the host machine. On the NVD, the packets are processed using a packet processing pipeline, which may include the execution of one or more VNICs and VRs associated with the VCN. For example, as part of the packet processing pipeline, the NVD executes or calls a function of a VNIC associated with the source compute instance (also referred to as executing the VNIC). The function executed by the VNIC may include checking a VLAN tag on the packet. Since the destination of the packet is outside the subnet, the VCN VR function is next called and executed by the NVD. The VCN VR then routes the packet to the NVD that executes the VNIC associated with the destination compute instance. The VNIC associated with the destination compute instance then processes the packet and forwards the packet to the destination compute instance. The VNICs associated with the source and destination compute instances may be executed on the same NVD (e.g., in the case where both the source and destination compute instances are hosted by the same host machine) or on different NVDs (e.g., in the case where the source and destination compute instances are hosted by different host machines connected to different NVDs).

[0125] If the destination for the packet is outside the VCN of the source compute instance, then the packet originating from the source compute instance is transmitted from the host machine hosting the source compute instance to the NVD connected to the host machine. The NVD executes the VNIC associated with the source compute instance. Because the destination endpoint of the packet is outside the VCN, the packet is then processed by the VCN VR of the VCN. The NVD invokes the VCN VR function, which may cause the packet to be forwarded to the NVD executing the appropriate gateway associated with the VCN. For example, if the destination is an endpoint within the customer's on-premises network, then the packet may be forwarded by the VCN VR to the NVD executing the DRG gateway configured for the VCN. The VCN VR may be executed on the same NVD as the NVD executing the VNIC associated with the source compute instance, or by a different NVD. The gateway may be executed by an NVD, which may be a smartNIC, a host machine, or other NVD implementation. The packet is then processed by the gateway and forwarded to the next hop, which facilitates delivery of the packet to its intended destination endpoint. For example, in Figure 2 , a packet originating from compute instance 268 may be transmitted from host machine 202 to NVD 210 via link 220 (using NIC 232). On NVD 210, VNIC 276 is invoked because it is the VNIC associated with source compute instance 268. VNIC 276 is configured to inspect information encapsulated in the packet and determine a next hop for forwarding the packet with the intent of facilitating delivery of the packet to its intended destination endpoint, and then forward the packet to the determined next hop.

[0126] Compute instances deployed on a VCN can communicate with a variety of different endpoints. These endpoints can include endpoints hosted by CSPI 200 and endpoints external to CSPI 200. Endpoints hosted by CSPI 200 can include instances in the same VCN or other VCNs, which can be the customer's VCN or a VCN that does not belong to the customer. Communication between endpoints hosted by CSPI 200 can be performed over physical network 218. Compute instances can also communicate with endpoints that are not hosted by CSPI 200 or endpoints external to CSPI 200. Examples of these endpoints include endpoints within a customer's on-premises deployment network or data center, or public endpoints accessible over a public network (such as the Internet). Communication with endpoints external to CSPI 200 can be performed over a public network (e.g., the Internet) using various communication protocols. Figure 2 not shown) or a private network ( Figure 2 (not shown) to execute.

[0127] Figure 2 The architecture of CSPI 200 depicted in the drawings is merely an example and is not intended to be limiting. In alternative embodiments, variations, alternatives, and modifications are possible. For example, in some embodiments, CSPI 200 may have a Figure 2 The systems or components shown in FIG. 5 may include more or fewer systems or components, may combine two or more systems, or may have different system configurations or arrangements. Figure 2 The systems, subsystems, and other components depicted in the may be implemented in software (e.g., code, instructions, programs) executed by one or more processing units (e.g., processors, cores) of the corresponding system, implemented using hardware, or implemented according to a combination thereof. The software may be stored on a non-transitory storage medium (e.g., on a memory device).

[0128] Figure 4 Depicted is a connection between a host machine and an NVD for providing I / O virtualization to support multi-tenancy in accordance with certain embodiments. Figure 4 As depicted in FIG. 4 , a host machine 402 executes a hypervisor 404 that provides a virtualized environment. The host machine 402 executes two virtual machine instances: VM1 406 belonging to customer / tenant #1 and VM2 408 belonging to customer / tenant #2. The host machine 402 includes a physical NIC 410 connected to an NVD 412 via a link 414. Each compute instance is attached to a VNIC executed by the NVD 412. Figure 4 In the embodiment of FIG. 4 , VM1 406 is attached to VNIC-VM1 420 and VM2 408 is attached to VNIC-VM2 422 .

[0129] like Figure 4 As shown in FIG. 4 , NIC 410 includes two logical NICs: logical NIC A 416 and logical NIC B 418. Each virtual machine is attached to and configured to work with its own logical NIC. For example, VM1 406 is attached to logical NIC A 416 and VM2 408 is attached to logical NIC B 418. Although host machine 402 includes only one physical NIC 410 shared by multiple tenants, each tenant's virtual machines believe they have their own host machine and NICs because of the logical NICs.

[0130] In some embodiments, each logical NIC is assigned its own VLAN ID. Thus, a particular VLAN ID is assigned to logical NIC A 416 for tenant #1, and a separate VLAN ID is assigned to logical NIC B 418 for tenant #2. When a packet is transmitted from VM1 406, a tag assigned to tenant #1 is attached to the packet by the hypervisor, and the packet is then transmitted from the host machine 402 to the NVD 412 via link 414. In a similar manner, when a packet is transmitted from VM2 408, a tag assigned to tenant #2 is attached to the packet by the hypervisor, and the packet is then transmitted from the host machine 402 to the NVD 412 via link 414. Thus, a packet 424 transmitted from the host machine 402 to the NVD 412 has an associated tag 426 that identifies the particular tenant and the associated VM. On the NVD, for a packet 424 received from the host machine 402, a tag 426 associated with the packet is used to determine whether the packet is processed by VNIC-VM1 420 or VNIC-VM2 422. The packet is then processed by the corresponding VNIC. Figure 4 The configuration described in enables each tenant's compute instances to believe that they have their own host machine and NIC. Figure 4 The setup depicted in provides I / O virtualization for supporting multi-tenancy.

[0131] Figure 5 A simplified block diagram of a physical network 500 is depicted in accordance with certain embodiments. Figure 5 The embodiments depicted in are structured as Clos networks. A Clos network is a specific type of network topology that is designed to provide connection redundancy while maintaining high bisection bandwidth and maximum resource utilization. A Clos network is a type of non-blocking, multi-stage or multi-layer switching network, where the number of stages or layers can be two, three, four, five, etc. Figure 5The embodiment depicted in is a 3-layer network, including layer 1, layer 2, and layer 3. TOR switch 504 represents a layer 0 switch in a Clos network. One or more NVDs are connected to the TOR switch. The layer 0 switch is also called the edge device of the physical network. The layer 0 switch is connected to the layer 1 switch, which is also called the leaf switch. Figure 5 In the embodiment depicted in , a set of "n" layer 0TOR switches is connected to a set of "n" layer 1 switches and together form a cluster (pod). Each layer 0 switch in a cluster is interconnected to all layer 1 switches in the cluster, but there is no connection between switches between clusters. In some embodiments, two clusters are referred to as blocks. Each block is served or connected to a set of "n" layer 2 switches (sometimes called backbone switches). There may be several blocks in the physical network topology. The layer 2 switches are in turn connected to "n" layer 3 switches (sometimes called super backbone switches). The communication of packets over the physical network 500 is typically performed using one or more layer 3 communication protocols. Typically, all layers of the physical network (except the TOR layer) are n-way redundant, thus allowing high availability. Policies can be specified for clusters and blocks to control the visibility of switches to each other in the physical network, thereby enabling the scaling of the physical network.

[0132] A Clos network is characterized by a fixed maximum number of hops from one layer 0 switch to another layer 0 switch (or from an NVD connected to a layer 0 switch to another NVD connected to a layer 0 switch). For example, in a 3-layer Clos network, a packet takes at most seven hops to reach from one NVD to another NVD, where the source and destination NVDs are connected to the leaf layer of the Clos network. Similarly, in a 4-layer Clos network, a packet takes at most nine hops to reach from one NVD to another NVD, where the source and destination NVDs are connected to the leaf layer of the Clos network. Therefore, the Clos network architecture maintains consistent latency throughout the network, which is important for communications within and between data centers. The Clos topology is horizontally scalable and cost-effective. The bandwidth / throughput capacity of the network can be easily increased by adding more switches at various layers (e.g., more leaf switches and spine switches) and by increasing the number of links between switches in adjacent layers.

[0133] In some embodiments, each resource within the CSPI is assigned a unique identifier, called a cloud identifier (CID). This identifier is included as part of the resource's information and can be used to manage the resource (e.g., via a console or through an API). An example syntax for a CID is:

[0134] ocidl.<RESOURCETPEREALM>.[REGION][.FUTUREUSE]<UNIQUEID>

[0135] in,

[0136] ocid1: a text string indicating the version of the CID;

[0137] resource type: the type of resource (for example, instance, volume, VCN, subnet, user, group, etc.);

[0138] realm: The realm where the resource is located. Example values ​​are "c1" for the commercial realm, "c2" for the government cloud realm, or "c3" for the federal government cloud realm. Each realm can have its own domain name;

[0139] region: The region where the resource is located. This part may be empty if the region does not apply to the resource;

[0140] future use: Reserved for future use.

[0141] unique ID: The unique portion of the ID. The format can vary depending on the type of resource or service.

[0142] Remote Direct Memory Access (RDMA) provides high throughput and ultra-low latency with low CPU overhead necessary for modern data center applications. RDMA is deployed on Ethernet using the RDMA over Converged Ethernet (RoCE) protocol, which relies on priority-based flow control (PFC) to achieve lossless (i.e., packet loss-free) networking. Data Center Quantized Congestion Notification (DC-QCN) is an end-to-end congestion control scheme for RoCE. DC-QCN is supported by combining Explicit Congestion Notification (ECN) and PFC to overcome the limitations of PFC to support end-to-end lossless networking.

[0143] Priority-based Flow Control (PFC) is a lossless transmission and congestion relief feature that works by providing fine-grained link-level flow control for each traffic flow (with priority) on a full-duplex Ethernet link. When the receive buffer on the switch interface fills to the threshold, the switch transmits a pause frame to the sender (the connected peer) to temporarily stop the sender from transmitting more frames. It should be noted that the buffer threshold must be low enough so that the sender has time to stop transmitting frames and the receiver can accept the frames already on the line before the buffer overflows. The switch automatically sets the queue buffer threshold to prevent frame loss.

[0144] According to some embodiments, when congestion forces one priority flow on a link to pause, all other priority flows on the link continue to send frames. Only frames of the paused priority are not transmitted. When the receive buffer empties below another threshold, the switch sends a message to start the flow again. However, depending on the amount of traffic on the link or the assigned priority, pausing traffic may cause congestion on the ingress port, which will eventually spread the congestion to the entire network.

[0145] Explicit Congestion Notification (ECN) enables end-to-end congestion notification between two endpoints. ECN is an IP-level flag used by protocols in higher layers (e.g., UDP and TCP). The two endpoints are an ECN-enabled sender and an ECN-enabled receiver. In a preferred embodiment, ECN is enabled on both endpoints and on all intermediate devices between the endpoints. In some embodiments where the intermediate devices do not support ECN (i.e., non-ECN devices), end-to-end ECN functionality may be degraded. It is worth noting that for non-ECN devices, congestion is detected entirely based on packets dropped by the non-ECN devices. ECN notifies the network of congestion with the goal of reducing packet loss and latency by causing the sending device to reduce the transmission rate until the congestion clears without dropping packets.

[0146] Data Center Quantized Congestion Notification (DC-QCN) is a combination of ECN and PFC to support end-to-end lossless Ethernet. ECN helps overcome the limitations of PFC to achieve lossless Ethernet. The idea behind DC-QCN is to allow ECN to perform flow control by reducing the transmission rate at the onset of congestion, thereby minimizing the time to trigger PFC, which would completely stop the flow. Correct operation of DC-QCN requires balancing two conflicting requirements: (1) ensuring that PFC is not triggered too early, i.e., before ECN has a chance to send congestion feedback to slow down the flow, and (2) ensuring that PFC is not triggered too late, causing packet loss due to buffer overflow. It should be noted that important parameters need to be correctly calculated and configured to achieve the above key requirements. Typically, workloads such as general processing unit (GPU) workloads cannot achieve the desired throughput level when executed on cluster networks. One reason for this problem is that RoCE congestion management running on RDMA network interface cards (NICs) is not optimized, i.e., the parameters of the various network-level devices and NICs on the host machine are not optimized to improve network throughput. A framework and mechanism are described below that allow a user to classify the type of traffic class for data packets (eg, RDMA traffic class) and handle each type of traffic class in a unique manner.

[0147] Go to Figure 6, illustrates an exemplary tree diagram illustrating different parameter settings according to some embodiments. By some embodiments, parameter settings may be categorized into two types or levels, namely, parameter settings performed at a network device level (i.e., a network device in a communication path between a source host machine and a destination host machine), and parameter settings performed at a network interface card (NIC) level (e.g., a NIC associated with a source host machine).

[0148] Parameters associated with the network device level may include parameters corresponding to ECN markings for different classes of RDMA traffic and the depth of queues (i.e., buffers) used in the network device. In some embodiments, RDMA traffic may be classified into high-performance database RDMA traffic (e.g., Oracle Exadata traffic), high-performance computing RDMA traffic (i.e., latency-sensitive type of RDMA traffic), and graphics processing unit (GPU) workload traffic (i.e., bandwidth-hungry type of RDMA traffic). As described later in reference Figure 8 As described, different parameters may be set to specific values ​​in order to handle each of the above types of RDMA traffic in a unique manner. With respect to queue depth, it may be appreciated that for bandwidth hungry RDMA traffic (e.g., GPU workload traffic), more buffers, i.e., a larger queue depth, may be used compared to latency sensitive RDMA traffic for which a smaller queue depth is used.

[0149] Regarding parameter settings performed at the network interface card (NIC) level (i.e., the NIC on the source host machine), the parameters to be set may include an adaptive retransmission parameter, a multi-queue pair parameter, a slow restart parameter, and a packet sequence number parameter. According to some embodiments, the adaptive retransmission parameter is a binary parameter (i.e., an ON / OFF parameter) that corresponds to the amount of time required for the sending host machine (or the receiving host machine) to recover in the event of an interruption on the other side (i.e., the receiving host machine (or the sending host machine). In some implementations, the adaptive retransmission parameter is set to ON. The multi-queue pair parameter corresponds to the number of queues used by the NIC when processing different classes of traffic. For example, according to some embodiments, for latency-sensitive RDMA traffic, the number of queue pairs is set to one, while for bandwidth-sensitive RDMA traffic, the number of queue pairs is set to four.

[0150] The slow restart parameter controls the rate at which the queue is built. Specifically, in some embodiments, the slow restart parameter is a binary parameter (i.e., an ON / OFF parameter) that is set to ON to avoid burst traffic. The packet sequence number (PSN) parameter corresponds to the number of packets that can be on the line (i.e., in transit) at any given moment. By some embodiments, the PSN parameter typically has a default value of 2. 7In some implementations, the PSN parameter is set to 2 9 It will be appreciated that setting the PSN parameter to a larger value (e.g., greater than 2 9 ) will result in a higher number of packet retransmissions to handle packet loss. Moreover, in some embodiments, the PSN parameter is made programmable, that is, the parameter is configurable so that it can be set to any desired value.

[0151] Now go to Figure 7 , illustrates an exemplary tree diagram illustrating different setting options for a Differentiated Services Code Point (DSCP) setting in accordance with certain embodiments. In some embodiments, a user is provided with the flexibility to mark (i.e., set) a portion of a header of a data packet to indicate a desired traffic type or traffic class to which the packet belongs. For example, in some embodiments, a source (e.g., an application executing on a source host machine) may be configured to set a portion of a header of a packet to indicate the type of traffic class to be associated with the packet. In doing so, the data packet is treated (i.e., processed) in a manner unique to the traffic class. For example, parameters at the network device level as well as parameters at the NIC level may be set in a manner so that data packets (belonging to a traffic class) are processed in an optimal manner.

[0152] By some embodiments, a header (e.g., an IP header) of a data packet may include a type of service (TOS) field that is one byte (i.e., 8 bits) in length. Two bits of the TOS field may be reserved for explicit congestion notification (ECN) purposes, while the remaining six bits may be used for differentiated services code point (DSCP) purposes. Thus, the DSCP bits may be marked in a particular manner to indicate the expected traffic type of the data packet and its corresponding treatment, e.g., how to handle ECN packets of a traffic class.

[0153] By one embodiment, the DSCP bit may be marked to indicate that the data packet belongs to a type of RDMA traffic that may be latency sensitive or bandwidth hungry. Thus, the ECN may be set by one of the following schemes: deterministic marking, probabilistic marking, a loose marking scheme, an aggressive marking scheme, or a combination thereof, i.e., deterministic and aggressive or loose and probabilistic. Figure 8 Such an ECN marking scheme is described. In addition, the DSCP bits can be utilized to indicate whether the data packet belongs to a lossy TCP type of traffic, or whether the data packet belongs to a DC-QCN managed type of traffic.

[0154] like Figure 7 As shown in , the first level of traffic classification can be obtained by marking the DSCP bits in a specific way, that is, setting the DSCP bits to indicate the type of traffic, such as delay-sensitive RDMA traffic, bandwidth-hungry RDMA traffic, etc. At the second level, multiple queue pairs (labeled as Q1-QK ) to process data packets of different types of RDMA traffic. For example, in one embodiment, for latency-sensitive RDMA traffic, a single queue may be used, while for bandwidth-hungry RDMA traffic (i.e., jitter-sensitive traffic), multiple queues, for example, four queues, may be used. It can be appreciated that a small number of queue pairs (e.g., one queue) minimizes the latency overhead in the NIC, while using multiple queue pairs (e.g., four queues) minimizes the impact of hash conflicts in the network. In addition, the third level of traffic classification corresponds to the way in which data packets of the traffic type are processed or executed. It can be appreciated that each type of traffic class may have different bandwidth requirements, for example, among multiple queues, the first queue may be allocated more bandwidth than the second queue, indicating that the first queue is expected to have a higher amount of traffic, while both queues require latency-sensitive treatment. In one embodiment, for latency-sensitive traffic, the threshold associated with the queue remains low and a deterministic marking scheme is adopted, while for bandwidth-hungry traffic, the threshold associated with the queue remains large and a probabilistic marking is adopted. Multiple queues may be used to obtain additional throughput for applications via parallelization. In addition, it should be noted that in some embodiments, multiple delay-sensitive traffic classes can be used to avoid scenarios where different types of delay-sensitive traffic cause queueing (and thus jitter) between each other. In other words, multiple delay-sensitive queues can be used to avoid head-of-line (HOL) blocking scenarios.

[0155] Figure 8 Depicted is a schematic diagram illustrating a mechanism for marking explicit congestion notification (ECN) packets according to certain embodiments. Specifically, Figure 8 Depicted is a diagram illustrating how a switch marks ECN packets for different types of RDMA traffic. Figure 8 As shown in FIG. 1 , the x-axis is labeled as the egress queue size of the switch, and the y-axis is labeled as the marking probability, i.e., the probability that a given packet will be marked as an ECN packet. According to some embodiments, the switch (e.g., TOR switch, backbone switch, etc.) performs ECN marking using the following three parameters: (1) a minimum threshold parameter (K) defined according to a segment or packet min ), (2) according to the maximum threshold parameter (K max ), and (3) a marking probability parameter. It should be noted that, according to some embodiments, the length of the segment can be 208 bytes.

[0156] In one embodiment, the switch monitors the depth of the buffer for each traffic class (i.e., the number of fragments present in the queue). When the buffer reaches the "minimum" ECN threshold (i.e., K min), the switch begins marking the ECN bits of packets. Marking can be performed in one of two ways: (1) in a random manner (i.e., probabilistic manner), where the probability of a particular packet being marked is defined by a configured marking probability, or (2) in a deterministic manner, where the probability of marking a packet is 100% based on satisfying a certain condition. Details regarding the manner in which packets are marked are described in further detail below.

[0157] In a probabilistic way of marking packets, the switch monitors the depth of the buffer for each traffic class (i.e., the number of fragments present in the queue). When the buffer reaches the “minimum” ECN threshold (i.e., K min ), the switch starts marking the ECN bit of the packet. When the buffer reaches the “maximum” ECN threshold (i.e., K max ),the switch has the maximum number of P max % of the packets are marked with ECN bits. In addition, in some embodiments, when the buffer depth exceeds the maximum ECN threshold (i.e., K max ), 100% of the packets are marked. min With K max The switch increases the percentage of marked packets linearly from 0% to P max %. In contrast, in the deterministic way of marking packets, the switch maintains only a single threshold, i.e., the switch maintains K min =K max = K. In deterministic marking, the switch monitors the depth of the buffer for each traffic class (i.e., the number of fragments present in the queue) and performs marking as follows: if the depth of the buffer is less than K, no packets are marked, and if the depth of the buffer is greater than K, all packets are marked with a probability of 100%. As shown below Fig. 9 As described, the switch parameters (e.g., P max , K min and K max ) can be set to a specific value based on the type of traffic handled by the switch to obtain optimal throughput performance.

[0158] Fig. 9 Depicted are diagrams illustrating different parameter settings (i.e., parameter P) for different types of RDMA traffic according to certain embodiments. max , K min and K max In one embodiment, when the type of traffic handled by the switch is bandwidth-optimized RDMA traffic, the switch may implement a "relaxed and probabilistic" mechanism for marking ECN packets. It will be appreciated that the relaxed feature is achieved by utilizing deeper (or larger) queue sizes in order to trade off a little latency (i.e., incur slightly higher latency) for higher throughput. By way of example, Fig. 9 As depicted in the table, for bandwidth-hungry RDMA traffic, the parameters can be set as follows: K min =2000, K max = 10,000 and probability P max =20%.

[0159] In addition, if Fig. 9 As shown in the table, for latency-sensitive RDMA applications, the switch can implement a “deterministic and aggressive” mechanism for marking ECN packets. Specifically, for this type of traffic (e.g., high-performance computing traffic), the parameters can be set as follows: K min =50, K max =50 (i.e., K min =K max =K) and the probability of marking P max =100%. Therefore, for this traffic type, the switch monitors the depth of the queue, and when it determines that the buffer exceeds K, the switch marks each subsequent packet with a probability of 100%. In addition, as mentioned earlier, multiple queue pairs can be used to obtain optimal performance. For example, for bandwidth-hungry RDMA traffic, multiple queues (e.g., four queues) can be used, while for latency-sensitive RDMA applications, a single queue can be used.

[0160] Furthermore, in the case where there are multiple queues for a certain type of traffic, priorities may be assigned to the different queues so as to achieve preferential treatment for at least some of the queues. For example, considering high performance database traffic (i.e., Exadata), four deterministic queues may be used, where traffic on the first queue (of the four queues) is assigned a higher priority than traffic on the other queues. By some embodiments, the priorities may correspond to bandwidth distribution of the queues, where the first queue is allocated 40% of the available bandwidth, while the remaining 60% of the bandwidth is distributed (e.g., in an even manner) among the other three queues.

[0161] According to some embodiments, the above-described specific parameter settings based on the type of traffic handled by the switch provide a 6x improvement in GPU workload performance, i.e., a six-fold improvement relative to using the default settings for all types of RDMA traffic. The following table illustrates the performance scaling achieved as the node count increases.

[0162]

[0163] Fig.10 A schematic diagram 1000 is depicted for identifying network locations in a cloud infrastructure to which different parameter tags are applied in accordance with certain embodiments. Fig.10 Depicts a portion of a network cluster. Fig.10As shown in , the network cluster may include one or more racks, for example, rack 1 1010. Each rack may include one or more host machines, for example, host 1-A 1012 and host 1-B 1014 included in rack 1 1010. Each host machine may include one or more NIC ports, for example, NIC port 1013 of host 1-A 1012 and NIC port 1015 of host 1-B 1014. Each NIC port of the host machine is connected to a host-facing port of a top of rack (TOR) switch, for example, TOR1 1016. TOR switch 1016 is in turn connected to a plurality of spine switches, for example, spine 1 1030 to spine P 1040.

[0164] According to some embodiments, the TOR switch 1016 and multiple spine switches (e.g., Spine 1 1030 and Spine P 1040) enable ECN marking 1071 as described above. Specifically, if there is any queue on any port of these switches, the switches execute the ECN marking scheme (for different traffic classes), as previously described with reference to Figure 8 as described. Moreover, the TOR switches and the backbone switches perform quality of service (QoS) marking. QoS marking enables the switch to determine which traffic class of data packets should be placed in which queue, i.e., QoS marking enables the switch to perform priority queue placement of data packets, for example, determining which data packet should be placed in which queue. It can be appreciated that such priority placement of data packets on corresponding queues can be performed based on the DSCP bits marked by the user on the terminal host. According to some embodiments, the TOR and the backbone switches are also configured to assign the highest priority to management queues (e.g., management queues carrying congestion notification packets (CNPs)), because such management queues manage the end-to-end performance of the network cluster.

[0165] In addition, if Fig.10 As shown in FIG, a priority flow control (PFC) setting 1072 is performed on the TOR port facing the host machine in the rack. It can be appreciated that the PFC setting operates at a per-queue level and is related to the amount of buffer capacity borrowed by the host NIC from the TOR switch. Furthermore, as Fig.10 As shown in , the host machine performs DSCP marking 1073 on the data packets transmitted by the host machine to the TOR switch. In addition, parameter settings related to adaptive retransmission, multiple queue pairs, slow restart and packet sequence number are performed on the NIC port of the host machine.

[0166] Fig.11A Illustrated is an exemplary flow chart 1100 depicting steps performed in processing a data packet based on its traffic class in accordance with certain embodiments. Fig.11AThe processing depicted in the can be implemented in software (e.g., code, instructions, program) executed by one or more processing units (e.g., processors, cores) of the corresponding system, hardware, or a combination thereof. The software can be stored on a non-transitory storage medium (e.g., on a memory device). Fig.11A The methods presented in and described below are intended to be illustrative and non-limiting. Fig.11A Various processing steps that occur in a particular order or sequence are depicted, but this is not intended to be limiting. In some alternative embodiments, these steps may be performed in a different order or some steps may be performed in parallel.

[0167] Processing begins at step 1105, where a network device in a communication path between a source host machine and a destination host machine obtains a data packet transmitted by the source host machine. Fig.10 , the source host machine may correspond to the host 1-B 1014 (included in rack 1 1010), and the destination host machine may correspond to another host machine included in another rack (not shown). In addition, the network device may correspond to the TOR 1 switch 1016.

[0168] In step 1110, the network device processes the received data packet. For example, the network device extracts a label included in a header portion of the data packet. The label contains information indicating the traffic type of the data packet. It can be appreciated that the label can be set by an application executed on the source host machine. For example, referring to Figure 7 , the DSCP bits included in the header portion of the data packet may be set in a specific manner to indicate a specific type of traffic to which the data packet belongs. According to some embodiments, the data packet may be one of several types of RDMA traffic, such as high-performance database traffic (e.g., exadata traffic), high-performance computing traffic (i.e., latency-sensitive type of RDMA traffic), GPU workload traffic (i.e., bandwidth-hungry type of RDMA traffic), RoCE v2 management type of traffic, lossy TCP traffic, etc. Then, the process moves to step 1115, where the network device determines the traffic class (of the data packet) based on the information included in the label.

[0169] In step 1120, the network device processes the data packet based on the traffic class determined in step 1115. For example, referring to Figure 6 , network device level parameters and / or parameters associated with the NIC on (one or more) host machines can be set to specific values ​​as described above in order to process data packets in a specific manner (i.e., based on the type of traffic class to which the data packets belong).

[0170] Fig. 11BAn exemplary flow chart 1130 depicting steps performed in processing data packets belonging to a traffic class of a bandwidth sensitive type is illustrated in accordance with certain embodiments. Fig. 11B The processing depicted in the can be implemented in software (e.g., code, instructions, programs), hardware, or a combination thereof, executed by one or more processing units (e.g., processors, cores) of the corresponding system. The software can be stored on a non-transitory storage medium (e.g., a memory device). Fig. 11B The methods presented in and described below are intended to be illustrative and non-limiting. Fig. 11B Various processing steps that occur in a particular order or sequence are depicted, but this is not intended to be limiting. In some alternative embodiments, these steps may be performed in a different order, or some steps may be performed in parallel.

[0171] Processing begins at step 1135, where a network device in the communication path between a source host machine and a destination host machine obtains a data packet transmitted by the source host machine. Fig.10 , the source host machine may correspond to the host 1-B 1014 (included in rack 1 1010), and the destination host machine may correspond to another host machine included in another rack (not shown). In addition, the network device may correspond to the TOR 1 switch 1016.

[0172] In step 1140, the network device processes the received data packet. For example, the network device may extract a label included in a header portion of the data packet. The label contains information indicating the traffic type of the data packet. It is appreciated that the label may be set by an application executed on the source host machine. For example, referring to Figure 7 , the user can set the DSCP bits included in the header portion of the data packet in a specific way to indicate the specific type of traffic to which the data packet belongs. According to some embodiments, the data packet can be one of several types of RDMA traffic, such as high-performance database traffic, high-performance computing (HPC) traffic (i.e., latency-sensitive type of RDMA traffic), GPU workload traffic (i.e., bandwidth-hungry type of RDMA traffic), RoCE v2 congestion control or congestion management type of traffic, lossy TCP traffic, etc. Then, the process moves to step 1145, where the network device determines that the data packet belongs to bandwidth-sensitive type of RDMA traffic based on the label.

[0173] In step 1147, the network device processes the data packet in a manner corresponding to processing bandwidth-sensitive type RDMA traffic. For example, one or more settings associated with bandwidth-sensitive type RDMA traffic are applied to multiple network devices involved in transmitting the data packet from the source host machine to the destination host machine. It can be appreciated that the one or more settings associated with bandwidth-sensitive type RDMA traffic can include processing the ECN marking of the data packet in a probabilistic and relaxed manner, for example, setting the value K min =2000 and K max =10000, the probability of marking a packet is set to 20%. In addition, since bandwidth-sensitive (i.e., bandwidth-hungry) applications can tolerate a certain amount of latency in exchange for higher bandwidth (i.e., higher throughput), a larger buffer size is used. It can be appreciated that by utilizing a larger buffer size, the threshold for how many data packets or fragments can be stored in the buffer is increased, for example, a buffer size of hundreds to thousands of fragments (e.g., where each fragment is 208 bytes long) can be utilized to handle bandwidth-sensitive types of RDMA traffic.

[0174] Fig. 11C An exemplary flow chart 1150 depicting steps performed in processing data packets belonging to a traffic class of a latency-sensitive type is illustrated in accordance with certain embodiments. Fig. 11C The processing depicted in the can be implemented in software (e.g., code, instructions, programs), hardware, or a combination thereof, executed by one or more processing units (e.g., processors, cores) of the corresponding system. The software can be stored on a non-transitory storage medium (e.g., a memory device). Fig. 11C The methods presented in and described below are intended to be illustrative and non-limiting. Fig. 11C Various processing steps that occur in a particular order or sequence are depicted, but this is not intended to be limiting. In some alternative embodiments, these steps may be performed in a different order, or some steps may be performed in parallel.

[0175] Processing begins at step 1151, where a network device in a communication path between a source host machine and a destination host machine obtains a data packet transmitted by the source host machine. Fig.10 , the source host machine may correspond to the host 1-B 1014 (included in rack 1 1010), and the destination host machine may correspond to another host included in another rack (not shown). In addition, the network device may correspond to the TOR 1 switch 1016.

[0176] In step 1153, the network device processes the received data packet. For example, the network device may extract a label included in a header portion of the data packet. The label contains information indicating the traffic type of the data packet. It is appreciated that the label may be set by an application executed on the source host machine. For example, referring to Figure 7 , the user can set the DSCP bits included in the header portion of the data packet in a specific way to indicate the specific type of traffic to which the data packet belongs. According to some embodiments, the data packet can be one of several types of RDMA traffic, such as high-performance database traffic, high-performance computing traffic (i.e., latency-sensitive type of RDMA traffic), GPU workload traffic (i.e., bandwidth-hungry type of RDMA traffic), DC-QCN management type of traffic, lossy TCP traffic, etc. Then, the process moves to step 1155, where the network device determines that the data packet belongs to latency-sensitive type of RDMA traffic based on the label.

[0177] In step 1157, the network device processes the data packet in a manner corresponding to processing the delay-sensitive type of RDMA traffic. For example, one or more settings associated with the delay-sensitive type of RDMA traffic are applied to multiple network devices involved in transmitting the data packet from the source host machine to the destination host machine. It can be appreciated that the one or more settings associated with the delay-sensitive type of RDMA traffic can include processing the ECN marking of the data packet in a deterministic and aggressive manner, for example, setting the value K min =50 and K max =50, the probability of marking a packet is set to 100%. Additionally, for delay-sensitive (i.e., jitter-sensitive) applications, a buffer size of tens of fragments (e.g., 50 fragments) may be used, where each fragment is 208 bytes long. It will be appreciated that by using a smaller buffer size, a lower threshold for how many data packets or fragments may be stored in the buffer, resulting in lower latency for the data packets. For example, by one embodiment, in response to determining that a data packet is of a delay-sensitive type, a parameter setting associated with the number of queues to be employed in (one or more) network devices in a communication path between a source host machine and a destination host machine may be set to one queue.

[0178] Fig.11D An exemplary flow chart 1170 depicting steps performed in tagging a portion of a header of a data packet is illustrated in accordance with certain embodiments. Fig.11D The processing depicted in the can be implemented in software (e.g., code, instructions, programs), hardware, or a combination thereof, executed by one or more processing units (e.g., processors, cores) of the corresponding system. The software can be stored on a non-transitory storage medium (e.g., a memory device). Fig.11DThe methods presented in and described below are intended to be illustrative and non-limiting. Fig.11D Various processing steps that occur in a particular order or sequence are depicted, but this is not intended to be limiting. In some alternative embodiments, these steps may be performed in a different order, or some steps may be performed in parallel.

[0179] Processing begins at step 1171, where a source host machine prepares data packets to be transmitted to a destination host machine. Fig.10 , a source host machine (e.g., host 1-B 1014 (included in rack 1 1010)) prepares a data packet to be transmitted to a destination host machine (e.g., another host included in another rack (not shown)). The process then moves to step 1173, where a portion of a header of the data packet is tagged (e.g., tagged by an application executing on the source host machine) to indicate an expected traffic type of the data packet and its corresponding processing. With some embodiments, the header of the data packet may include a type of service (TOS) field having a length of one byte. Two bits of the TOS field may be reserved for explicit congestion notification (ECN) purposes, while the remaining six bits may be used for differentiated services code point (DSCP) purposes. Thus, the DSCP bits may be marked in a particular manner to indicate an expected traffic type of the data packet and its corresponding processing, for example, how to handle ECN packets of a traffic class.

[0180] After tagging a portion of the header of the data packet (e.g., the DSCP bits of the TOS field), the process in step 1175 forwards the data packet to the destination host machine on the communication path by the source host machine. Fig.10 , the source host machine (e.g., host 1-B 1014) can forward the data packet to the TOR1 switch 1016.

[0181] Example Cloud Infrastructure Embodiment

[0182] As mentioned above, Infrastructure as a Service (IaaS) is a specific type of cloud computing. IaaS can be configured to provide virtualized computing resources over a public network (e.g., the Internet). In the IaaS model, a cloud computing provider can host infrastructure components (e.g., servers, storage devices, network nodes (e.g., hardware), deployment software, platform virtualization (e.g., hypervisor layer), etc.). In some cases, IaaS providers can also provide various services to accompany these infrastructure components (e.g., billing, monitoring, logging, security, load balancing and clustering, etc.). Therefore, since these services may be policy-driven, IaaS users may be able to implement policies to drive load balancing to maintain application availability and performance.

[0183] In some cases, IaaS customers can access resources and services over a wide area network (WAN) such as the Internet, and can use the cloud provider's services to install the remaining elements of the application stack. For example, a user can log in to the IaaS platform to create a virtual machine (VM), install an operating system (OS) on each VM, deploy middleware such as a database, create storage buckets for workloads and backups, and even install enterprise software into the VM. Customers can then use the provider's services to perform a variety of functions, including balancing network traffic, troubleshooting application problems, monitoring performance, managing disaster recovery, and more.

[0184] In most cases, the cloud computing model will require the involvement of a cloud provider. A cloud provider may, but need not, specialize in providing (e.g., provisioning, renting, selling) third-party IaaS services. An entity may also choose to deploy a private cloud, thereby becoming its own infrastructure service provider.

[0185] In some examples, IaaS deployment is the process of placing a new application or a new version of an application onto a prepared application server, etc. It may also include the process of preparing the server (e.g., installing libraries, daemons, etc.). This is typically managed by the cloud provider, below the hypervisor layer (e.g., servers, storage, network hardware, and virtualization). Thus, the customer may be responsible for handling (OS), middleware, and / or application deployment (e.g., on self-service virtual machines (e.g., that can be started on demand)), etc.

[0186] In some examples, IaaS provisioning can refer to obtaining computers or virtual hosts for use, and even installing required libraries or services on them. In most cases, deployment does not include provisioning, and provisioning may need to be performed first.

[0187] In some cases, there are two different challenges with IaaS provisioning. First, there is the initial challenge of provisioning the initial set of infrastructure before anything is running. Second, once everything has been provisioned, there is the challenge of evolving the existing infrastructure (e.g., adding new services, changing services, removing services, etc.). In some cases, both of these challenges can be addressed by enabling configuration that defines the infrastructure in a declarative manner. In other words, the infrastructure (e.g., which components are required and how they interact) can be defined by one or more configuration files. Therefore, the overall topology of the infrastructure (e.g., which resources depend on which resources, and how they work together) can be described in a declarative manner. In some cases, once the topology is defined, workflows can be generated to create and / or manage the different components described in the configuration files.

[0188] In some examples, the infrastructure may have many interconnected elements. For example, there may be one or more virtual private clouds (VPCs) (e.g., a potential on-demand pool of configurable and / or shared computing resources), also referred to as a core network. In some examples, one or more security group rules may also be provided to define how the security of the network will be set up and one or more virtual machines (VMs). Other infrastructure elements may also be provided, such as load balancers, databases, etc. The infrastructure may evolve gradually as more and more infrastructure elements are desired and / or added.

[0189] In some cases, continuous deployment techniques can be employed to enable infrastructure code to be deployed across various virtual computing environments. In addition, the described techniques can enable infrastructure management within these environments. In some examples, a service team can write code that is expected to be deployed to one or more but typically many different production environments (e.g., across various different geographic locations, sometimes across the world). However, in some examples, the infrastructure on which the code will be deployed must first be set up. In some cases, provisioning can be done manually, resources can be provisioned using a provisioning tool, and / or code can be deployed using a deployment tool once the infrastructure is provisioned.

[0190] Fig.12 1 is a block diagram 1200 illustrating an example mode of an IaaS architecture according to at least one embodiment. A service operator 1202 can be communicatively coupled to a secure host lease 1204 that can include a virtual cloud network (VCN) 1206 and a secure host subnet 1208. In some examples, the service operator 1202 can use one or more client computing devices (the client computing device can be a portable handheld device (e.g., Cellular phone, computing tablets, personal digital assistants (PDAs), or wearable devices (e.g., Google head mounted display)), running software (such as Microsoft Windows ) and / or various mobile operating systems (such as iOS, Windows Phone, Android, BlackBerry 8, Palm OS, etc.), and supports Internet, email, short message service (SMS), Alternatively, the client computing device may be a general-purpose personal computer, including, for example, a computer running various versions of Microsoft Apple The client computing device may be a personal computer and / or laptop computer running various commercially available or any workstation computer running any operating system such as UNIX or UNIX-like operating systems, including but not limited to various GNU / Linux operating systems (such as, for example, Google Chrome OS). Alternatively or additionally, the client computing device may be any other electronic device, such as a thin client computer, an Internet-enabled gaming system (e.g., with or without gesture input device), and / or a personal messaging device capable of communicating over a network that can access VCN 1206 and / or the Internet.

[0191] VCN 1206 may include a local peer gateway (LPG) 1210, which may be communicatively coupled to a secure shell (SSH) VCN 1212 via LPG 1210 contained in SSH VCN 1212. SSH VCN 1212 may include an SSH subnet 1214, and SSH VCN 1212 may be communicatively coupled to a control plane VCN 1216 via LPG 1210 contained in control plane VCN 1216. In addition, SSH VCN 1212 may be communicatively coupled to a data plane VCN 1218 via LPG 1210. Control plane VCN 1216 and data plane VCN 1218 may be contained in a service lease 1219, which may be owned and / or operated by an IaaS provider.

[0192] The control plane VCN 1216 may include a control plane demilitarized zone (DMZ) layer 1220 that acts as a perimeter network (e.g., a portion of a company network between a company's internal network and an external network). DMZ-based servers can assume limited liability and help control security violations. In addition, the DMZ layer 1220 may include one or more load balancer (LB) subnets 1222, a control plane application layer 1224 that may include (one or more) application (app) subnets 1226, and a control plane data layer 1228 that may include (one or more) database (DB) subnets 1230 (e.g., (one or more) front-end DB subnets and / or (one or more) back-end DB subnets). The LB subnet(s) 1222 included in the control plane DMZ layer 1220 may be communicatively coupled to the application subnet(s) 1226 included in the control plane application layer 1224 and the Internet gateway 1234 that may be included in the control plane VCN 1216, and the application subnet(s) 1226 may be communicatively coupled to the DB subnet(s) 1230 included in the control plane data layer 1228 and the service gateway 1236 and the network address translation (NAT) gateway 1238. The control plane VCN 1216 may include the service gateway 1236 and the NAT gateway 1238.

[0193] The control plane VCN 1216 may include a data plane mirror application layer 1240, which may include application subnet(s) 1226. The application subnet(s) 1226 included in the data plane mirror application layer 1240 may include a virtual network interface controller (VNIC) 1242 that may execute a compute instance 1244. The compute instance 1244 may communicatively couple the application subnet(s) 1226 of the data plane mirror application layer 1240 to the application subnet(s) 1226 that may be included in the data plane application layer 1246.

[0194] The data plane VCN 1218 may include a data plane application layer 1246, a data plane DMZ layer 1248, and a data plane data layer 1250. The data plane DMZ layer 1248 may include (one or more) LB subnets 1222, which may be communicatively coupled to (one or more) application subnets 1226 of the data plane application layer 1246 and an internet gateway 1234 of the data plane VCN 1218. The (one or more) application subnets 1226 may be communicatively coupled to a service gateway 1236 of the data plane VCN 1218 and a NAT gateway 1238 of the data plane VCN 1218. The data plane data layer 1250 may also include (one or more) DB subnets 1230, which may be communicatively coupled to (one or more) application subnets 1226 of the data plane application layer 1246.

[0195] The internet gateway 1234 of the control plane VCN 1216 and the data plane VCN 1218 may be communicatively coupled to a metadata management service 1252, which may be communicatively coupled to a public Internet 1254. The public Internet 1254 may be communicatively coupled to a NAT gateway 1238 of the control plane VCN 1216 and the data plane VCN 1218. The service gateway 1236 of the control plane VCN 1216 and the data plane VCN 1218 may be communicatively coupled to a cloud service 1256.

[0196] In some examples, service gateway 1236 of control plane VCN 1216 or data plane VCN 1218 can make application programming interface (API) calls to cloud services 1256 without going through public internet 1254. API calls from service gateway 1236 to cloud services 1256 can be one-way: service gateway 1236 can make API calls to cloud services 1256, and cloud services 1256 can send requested data to service gateway 1236. However, cloud services 1256 may not initiate API calls to service gateway 1236.

[0197] In some examples, secure host lease 1204 can be directly connected to service lease 1219, which can otherwise be isolated. Secure host subnet 1208 can communicate with SSH subnet 1214 through LPG 1210, which can enable two-way communication on otherwise isolated systems. Connecting secure host subnet 1208 to SSH subnet 1214 can enable secure host subnet 1208 to access other entities within service lease 1219.

[0198] The control plane VCN 1216 may allow a user of the service lease 1219 to set up or otherwise provision desired resources. The desired resources provisioned in the control plane VCN 1216 may be deployed or otherwise used in the data plane VCN 1218. In some examples, the control plane VCN 1216 may be isolated from the data plane VCN 1218, and the data plane mirror application layer 1240 of the control plane VCN 1216 may communicate with the data plane application layer 1246 of the data plane VCN 1218 via the VNIC 1242, which may be included in the data plane mirror application layer 1240 and the data plane application layer 1246.

[0199] In some examples, a user or customer of the system may make a request, such as a create, read, update, or delete (CRUD) operation, through the public internet 1254 which may transmit the request to the metadata management service 1252. The metadata management service 1252 may transmit the request to the control plane VCN 1216 through the internet gateway 1234. The request may be received by the LB subnet(s) 1222 contained in the control plane DMZ layer 1220. The LB subnet(s) 1222 may determine that the request is valid, and in response to the determination, the LB subnet(s) 1222 may transmit the request to the application subnet(s) 1226 contained in the control plane application layer 1224. If the request is validated and a call to the public internet 1254 is required, the call to the public internet 1254 may be transmitted to the NAT gateway 1238 which may make the call to the public internet 1254. The storage that the request may desire to be stored may be stored in the DB subnet(s) 1230.

[0200] In some examples, data plane mirror application layer 1240 may facilitate direct communications between control plane VCN 1216 and data plane VCN 1218. For example, it may be desirable to apply changes, updates, or other appropriate modifications to the configuration of resources contained in data plane VCN 1218. Control plane VCN 1216, via VNIC 1242, may communicate directly with the resources contained in data plane VCN 1218 and may thereby perform changes, updates, or other appropriate modifications to the configuration of these resources.

[0201] In some embodiments, control plane VCN 1216 and data plane VCN 1218 may be included in service lease 1219. In this case, a user or customer of the system may not own or operate control plane VCN 1216 or data plane VCN 1218. Alternatively, an IaaS provider may own or operate control plane VCN 1216 and data plane VCN 1218, both of which may be included in service lease 1219. This embodiment may enable isolation of networks that may prevent a user or customer from interacting with the resources of other users or other customers. In addition, this embodiment may allow a user or customer of the system to store a database privately without relying on the public Internet 1254, which may not have the desired level of security, for storage.

[0202] In other embodiments, the LB subnet(s) 1222 contained in the control plane VCN 1216 can be configured to receive signals from the service gateway 1236. In this embodiment, the control plane VCN 1216 and the data plane VCN 1218 can be configured to be called by customers of the IaaS provider without calling the public Internet 1254. Customers of the IaaS provider may desire this embodiment because the database(s) used by the customer can be controlled by the IaaS provider and can be stored on the service lease 1219, which can be isolated from the public Internet 1254.

[0203] Fig.13 13 is a block diagram 1300 illustrating another example model of an IaaS architecture according to at least one embodiment. A service operator 1302 (e.g., Fig.12 The service operator 1202 of the embodiment may be communicatively coupled to the secure host lease 1304 (e.g., Fig.12 The secure host lease 1204 may include a virtual cloud network (VCN) 1306 (e.g., Fig.12 VCN 1206) and secure host subnet 1308 (e.g., Fig.12 The VCN 1306 may include a local peer gateway (LPG) 1310 (e.g., Fig.12 LPG 1210), the VCN 1306 may be connected to the VCN 1312 (e.g., Fig.12 LPG 1210 in SSH VCN 1212 of the present invention is communicatively coupled to SSH VCN 1312. SSH VCN 1312 may include SSH subnet 1314 (e.g., Fig.12 SSH subnet 1214), and SSH VCN 1312 can be accessed via control plane VCN 1316 (e.g., Fig.12 LPG 1310 in control plane VCN 1216 of the present invention is communicatively coupled to control plane VCN 1316. Control plane VCN 1316 may be included in service lease 1319 (e.g., Fig.12 service lease 1219), and the data plane VCN 1318 (e.g., Fig.12 The data plane VCN 1218) may be contained in a customer lease 1321 which may be owned or operated by a user or customer of the system.

[0204] The control plane VCN 1316 may include a control plane DMZ layer 1320 (e.g., Fig.12, which may include (one or more) LB subnets 1322 (e.g., Fig.12 (one or more) LB subnets 1222), may include (one or more) application subnets 1326 (e.g., Fig.12 (one or more) application subnets 1226) of the control plane application layer 1324 (e.g., Fig.12 The control plane application layer 1224 of FIG. 1224 may include (one or more) database (DB) subnets 1330 (e.g., similar to Fig.12 (one or more) DB subnets 1230) of the control plane data layer 1328 (e.g., Fig.12 1328). The LB subnet(s) 1322 contained in the control plane DMZ layer 1320 may be communicatively coupled to the application subnet(s) 1326 contained in the control plane application layer 1324 and the internet gateway 1334 (e.g., Fig.12 1234), and the application subnet(s) 1326 may be communicatively coupled to the DB subnet(s) 1330 and the service gateway 1336 (e.g., Fig.12 Serving gateway) and network address translation (NAT) gateway 1338 (e.g., Fig.12 The control plane VCN 1316 may include a service gateway 1336 and a NAT gateway 1338.

[0205] The control plane VCN 1316 may include a data plane mirror application layer 1340 (e.g., Fig.12 The data plane image application layer 1240). The (one or more) application subnets 1326 included in the data plane image application layer 1340 may include a computing instance 1344 (e.g., similar to Fig.12 The compute instance 1344 can facilitate the application subnet(s) 1326 of the data plane mirroring application layer 1340 and can include the application subnets 1326 of the data plane mirroring application layer 1346 (e.g., Fig.12 Communication between (one or more) application subnets 1326 in the data plane application layer 1246) via the VNIC 1342 contained in the data plane mirror application layer 1340 and the VNIC 1342 contained in the data plane application layer 1346.

[0206] The Internet gateway 1334 included in the control plane VCN 1316 can be communicatively coupled to the metadata management service 1352 (e.g., Fig.12 1352), which can be communicatively coupled to the public Internet 1354 (e.g., Fig.12 The public Internet 1354 can be communicatively coupled to a NAT gateway 1338 included in the control plane VCN 1316. The service gateway 1336 included in the control plane VCN 1316 can be communicatively coupled to cloud services 1356 (e.g., Fig.12 Cloud service 1256).

[0207] In some examples, the data plane VCN 1318 may be contained in a customer lease 1321. In this case, the IaaS provider may provide a control plane VCN 1316 for each customer, and the IaaS provider may provision each customer with a unique compute instance 1344 contained in a service lease 1319. Each compute instance 1344 may allow communication between a control plane VCN 1316 contained in a service lease 1319 and a data plane VCN 1318 contained in a customer lease 1321. The compute instance 1344 may allow resources provisioned in a control plane VCN 1316 contained in a service lease 1319 to be deployed or otherwise used in a data plane VCN 1318 contained in a customer lease 1321.

[0208] In other examples, a customer of the IaaS provider may have a database that exists in the customer tenancy 1321. In this example, the control plane VCN 1316 may include a data plane mirror application layer 1340, which may include (one or more) application subnets 1326. The data plane mirror application layer 1340 may reside in the data plane VCN 1318, but the data plane mirror application layer 1340 may not be in the data plane VCN 1318. In other words, the data plane mirror application layer 1340 may access the customer tenancy 1321, but the data plane mirror application layer 1340 may not exist in the data plane VCN 1318 or be owned or operated by the customer of the IaaS provider. The data plane mirror application layer 1340 may be configured to make calls to the data plane VCN 1318, but may not be configured to make calls to any entity contained in the control plane VCN 1316. A customer may desire to deploy or otherwise use resources provisioned in control plane VCN 1316 in data plane VCN 1318, and data plane mirror application layer 1340 may facilitate the customer's desired deployment or other use of resources.

[0209] In some embodiments, a customer of an IaaS provider may apply filters to the data plane VCN 1318. In this embodiment, the customer may determine what the data plane VCN 1318 may access, and the customer may restrict access to the public Internet 1354 from the data plane VCN 1318. The IaaS provider may not apply filters or otherwise control access to any external network or database by the data plane VCN 1318. The customer applying filters and controls to the data plane VCN 1318 contained in the customer lease 1321 may help isolate the data plane VCN 1318 from other customers and the public Internet 1354.

[0210] In some embodiments, cloud services 1356 may be called by service gateway 1336 to access services that may not exist on public Internet 1354, control plane VCN 1316, or data plane VCN 1318. The connection between cloud services 1356 and control plane VCN 1316 or data plane VCN 1318 may not be real-time or continuous. Cloud services 1356 may exist on different networks owned or operated by IaaS providers. Cloud services 1356 may be configured to receive calls from service gateway 1336 and may be configured not to receive calls from public Internet 1354. Some cloud services 1356 may be isolated from other cloud services 1356, and control plane VCN 1316 may be isolated from cloud services 1356 that may not be in the same region as control plane VCN 1316. For example, control plane VCN 1316 may be located in "region 1", and cloud service "deployment 12" may be located in region 1 and "region 2". If a service gateway 1336 contained in a control plane VCN 1316 located in region 1 makes a call to a deployment 12, the call may be transmitted to the deployment 12 in region 1. In this example, the control plane VCN 1316 or deployment 12 in region 1 may not be communicatively coupled or otherwise in communication with the deployment 12 in region 2.

[0211] Fig.14 1400 is a block diagram illustrating another example model of an IaaS architecture according to at least one embodiment. A service operator 1402 (e.g., Fig.12 The service operator 1202 may be communicatively coupled to the secure host lease 1404 (e.g., Fig.12 The secure host lease 1204 may include a virtual cloud network (VCN) 1406 (e.g., Fig.12 VCN 1206) and secure host subnet 1408 (e.g., Fig.12 VCN 1406 may include LPG 1410 (e.g., Fig.12 LPG 1210), the VCN 1406 can be connected to the SSH VCN 1412 (e.g., Fig.12 LPG 1410 in SSH VCN 1212 of the present invention is communicatively coupled to SSH VCN 1412. SSH VCN 1412 may include SSH subnet 1414 (e.g., Fig.12 SSH subnet 1214), and SSH VCN 1412 can be accessed via control plane VCN 1416 (e.g., Fig.12 LPG 1410 in control plane VCN 1216 of the present invention is communicatively coupled to control plane VCN 1416 and communicates with control plane VCN 1416 via LPG 1410 included in data plane VCN 1418 (e.g., Fig.12 The control plane VCN 1416 and the data plane VCN 1418 may be included in a service lease 1419 (e.g., Fig.12 Service lease 1219).

[0212] The control plane VCN 1416 may include a load balancer (LB) subnet 1422 (e.g., Fig.12 The control plane DMZ layer 1420 (e.g., Fig.12 The control plane DMZ layer 1220 may include (one or more) application subnets 1426 (e.g., similar to Fig.12 (one or more) application subnets 1226) of the control plane application layer 1424 (e.g., Fig.12 A control plane application layer 1224), a control plane data layer 1428 (eg, Fig.12 1428). The LB subnet(s) 1422 contained in the control plane DMZ layer 1420 may be communicatively coupled to the application subnet(s) 1426 contained in the control plane application layer 1424 and the internet gateway 1434 (e.g., Fig.12 1434), and the application subnet(s) 1426 may be communicatively coupled to the DB subnet(s) 1430 and the service gateway 1436 (e.g., Fig.12 Serving gateway) and network address translation (NAT) gateway 1438 (e.g., Fig.12The control plane VCN 1416 may include a service gateway 1436 and a NAT gateway 1438.

[0213] The data plane VCN 1418 may include a data plane application layer 1446 (e.g., Fig.12 Data plane application layer 1246), data plane DMZ layer 1448 (e.g., Fig.12 Data plane DMZ layer 1248), and data plane data layer 1450 (e.g., Fig.12 The data plane DMZ layer 1448 may include (one or more) trusted application subnets 1460 and (one or more) untrusted application subnets 1462 that may be communicatively coupled to the data plane application layer 1446 and (one or more) LB subnets 1422 of the Internet gateway 1434 included in the data plane VCN 1418. The (one or more) trusted application subnets 1460 may be communicatively coupled to the service gateway 1436 included in the data plane VCN 1418, the NAT gateway 1438 included in the data plane VCN 1418, and (one or more) DB subnets 1430 included in the data plane data layer 1450. The (one or more) untrusted application subnets 1462 may be communicatively coupled to the service gateway 1436 included in the data plane VCN 1418 and (one or more) DB subnets 1430 included in the data plane data layer 1450. Data plane data layer 1450 may include DB subnet(s) 1430 that may be communicatively coupled to a service gateway 1436 included in data plane VCN 1418 .

[0214] Untrusted application subnet(s) 1462 may include one or more primary VNICs 1464(1)-(N) that may be communicatively coupled to tenant virtual machines (VMs) 1466(1)-(N). Each tenant VM 1466(1)-(N) may be communicatively coupled to a respective application subnet 1467(1)-(N) that may be contained in a respective container egress VCN 1468(1)-(N), which may be contained in a respective customer tenancy 1470(1)-(N). Respective secondary VNICs 1472(1)-(N) may facilitate communications between untrusted application subnet(s) 1462 contained in data plane VCN 1418 and application subnets contained in container egress VCNs 1468(1)-(N). Each container egress VCN 1468(1)-(N) may include a NAT gateway 1438 that may be communicatively coupled to the public Internet 1454 (e.g., Fig.12 of the public Internet 1254).

[0215] The Internet gateway 1434 included in the control plane VCN 1416 and included in the data plane VCN 1418 can be communicatively coupled to the metadata management service 1452 (e.g., Fig.12 1452), which can be communicatively coupled to the public Internet 1454. The public Internet 1454 can be communicatively coupled to the NAT gateway 1438 included in the control plane VCN 1416 and included in the data plane VCN 1418. The service gateway 1436 included in the control plane VCN 1416 and included in the data plane VCN 1418 can be communicatively coupled to the cloud service 1456.

[0216] In some embodiments, data plane VCN 1418 may be integrated with customer lease 1470. In some cases, such integration may be useful or desirable for customers of the IaaS provider, such as where support may be desired when executing code. A customer may provide code to run that may be destructive, may communicate with other customer resources, or may otherwise cause undesirable effects. In response to this, the IaaS provider may determine whether to run code given to the IaaS provider by the customer.

[0217] In some examples, a customer of an IaaS provider may grant temporary network access to the IaaS provider and request functionality attached to a data plane layer application 1446. Code running the functionality may be executed in a VM 1466(1)-(N), and the code may not be configured to run anywhere else on the data plane VCN 1418. Each VM 1466(1)-(N) may be connected to a customer tenancy 1470. The corresponding container 1471(1)-(N) contained in the VM 1466(1)-(N) may be configured to run the code. In this case, there may be double isolation (e.g., the container 1471(1)-(N) runs the code, where the container 1471(1)-(N) may be contained at least in the VM 1466(1)-(N) contained in the (one or more) untrusted application subnets 1462), which may help prevent incorrect or otherwise undesirable code from damaging the IaaS provider's network or damaging the network of a different customer. Containers 1471(1)-(N) may be communicatively coupled to customer lease 1470 and may be configured to transmit or receive data from customer lease 1470. Containers 1471(1)-(N) may not be configured to transmit or receive data from any other entity in data plane VCN 1418. After running the code is complete, the IaaS provider may terminate or otherwise dispose of containers 1471(1)-(N).

[0218] In some embodiments, the trusted application subnet(s) 1460 may run code that may be owned or operated by the IaaS provider. In this embodiment, the trusted application subnet(s) 1460 may be communicatively coupled to the DB subnet(s) 1430 and configured to perform CRUD operations in the DB subnet(s) 1430. The untrusted application subnet(s) 1462 may be communicatively coupled to the DB subnet(s) 1430, but in this embodiment, the untrusted application subnet(s) may be configured to perform read operations in the DB subnet(s) 1430. The containers 1471(1)-(N) that may be included in each customer's VM 1466(1)-(N) and that may run code from the customer may not be communicatively coupled to the DB subnet(s) 1430.

[0219] In other embodiments, the control plane VCN 1416 and the data plane VCN 1418 may not be directly communicatively coupled. In this embodiment, there may be no direct communication between the control plane VCN 1416 and the data plane VCN 1418. However, communication may occur indirectly through at least one method. The LPG 1410 may be established by an IaaS provider, which may facilitate communication between the control plane VCN 1416 and the data plane VCN 1418. In another example, the control plane VCN 1416 or the data plane VCN 1418 may make a call to the cloud service 1456 via the service gateway 1436. For example, the call from the control plane VCN 1416 to the cloud service 1456 may include a request for a service that can communicate with the data plane VCN 1418.

[0220] Fig.15 1500 is a block diagram illustrating another example model of an IaaS architecture according to at least one embodiment. A service operator 1502 (e.g., Fig.12 The service operator 1202 of the embodiment may be communicatively coupled to the secure host lease 1504 (e.g., Fig.12 The secure host lease 1204 may include a virtual cloud network (VCN) 1506 (e.g., Fig.12 VCN 1206) and secure host subnet 1508 (e.g., Fig.12 VCN 1506 may include LPG 1510 (e.g., Fig.12 1210), the VCN 1506 may be connected to the LPG 1210 via the SSH VCN 1512 (e.g., Fig.12LPG 1510 in SSH VCN 1512 of the present invention is communicatively coupled to SSH VCN 1512. SSH VCN 1512 may include SSH subnet 1514 (e.g., Fig.12 SSH subnet 1214), and SSH VCN 1512 can be accessed via control plane VCN 1516 (e.g., Fig.12 LPG 1510 in control plane VCN 1216 of the present invention is communicatively coupled to control plane VCN 1516 and communicates with control plane VCN 1516 via LPG 1510 included in data plane VCN 1518 (e.g., Fig.12 The control plane VCN 1516 and the data plane VCN 1518 may be included in a service lease 1519 (e.g., Fig.12 Service lease 1219).

[0221] The control plane VCN 1516 may include (one or more) LB subnets 1522 (e.g., Fig.12 The control plane DMZ layer 1520 (e.g., Fig.12 The control plane DMZ layer 1220 may include (one or more) application subnets 1526 (e.g., Fig.12 (one or more) application subnets 1226) of the control plane application layer 1524 (e.g., Fig.12 The control plane application layer 1224 of FIG. 1224 may include (one or more) DB subnets 1530 (e.g., Fig.14 (one or more) DB subnets 1430) of the control plane data layer 1528 (e.g., Fig.12 1528). The LB subnet(s) 1522 contained in the control plane DMZ layer 1520 may be communicatively coupled to the application subnet(s) 1526 contained in the control plane application layer 1524 and the internet gateway 1534 (e.g., Fig.12 1234), and the application subnet(s) 1526 may be communicatively coupled to the DB subnet(s) 1530 and the service gateway 1536 (e.g., Fig.12 ) and a network address translation (NAT) gateway 1538 (e.g., Fig.12 The control plane VCN 1516 may include a service gateway 1536 and a NAT gateway 1538.

[0222] The data plane VCN 1518 may include a data plane application layer 1546 (e.g., Fig.12 Data plane application layer 1246), data plane DMZ layer 1548 (e.g., Fig.12 Data plane DMZ layer 1248)), and data plane data layer 1550 (e.g., Fig.12 The data plane DMZ layer 1548 may include (one or more) trusted application subnets 1560 (e.g., Fig.14 (one or more) trusted application subnets 1460) and (one or more) untrusted application subnets 1562 (e.g., Fig.14 The trusted application subnet(s) 1560 may be communicatively coupled to the service gateway 1536 contained in the data plane VCN 1518, the NAT gateway 1538 contained in the data plane VCN 1518, and the DB subnet(s) 1530 contained in the data plane data layer 1550. The untrusted application subnet(s) 1562 may be communicatively coupled to the service gateway 1536 contained in the data plane VCN 1518 and the DB subnet(s) 1530 contained in the data plane data layer 1550. The data plane data layer 1550 may include the DB subnet(s) 1530 that may be communicatively coupled to the service gateway 1536 contained in the data plane VCN 1518.

[0223] The untrusted application subnet(s) 1562 may include primary VNICs 1564(1)-(N) that may be communicatively coupled to tenant virtual machines (VMs) 1566(1)-(N) residing within the untrusted application subnet(s) 1562. Each tenant VM 1566(1)-(N) may run code in a respective container 1567(1)-(N) and may be communicatively coupled to an application subnet 1526 that may be contained in a data plane application layer 1546, which may be contained in a container egress VCN 1568. A respective secondary VNIC 1572(1)-(N) may facilitate communications between the untrusted application subnet(s) 1562 contained in the data plane VCN 1518 and an application subnet contained in the container egress VCN 1568. The container egress VCN may include a primary VNIC 1564(1)-(N) that may be communicatively coupled to a public Internet 1554 (e.g., Fig.12 NAT gateway 1538 of the public Internet 1254).

[0224] The Internet gateway 1534 included in the control plane VCN 1516 and included in the data plane VCN 1518 can be communicatively coupled to the metadata management service 1552 (e.g., Fig.12 15), which can be communicatively coupled to the public Internet 1554. The public Internet 1554 can be communicatively coupled to the NAT gateway 1538 included in the control plane VCN 1516 and included in the data plane VCN 1518. The service gateway 1536 included in the control plane VCN 1516 and included in the data plane VCN 1518 can be communicatively coupled to the cloud service 1556.

[0225] In some examples, Fig.15 The architecture model shown in block diagram 1500 can be considered as Fig.14 1400, and may be desirable to customers of the IaaS provider in situations where the IaaS provider cannot communicate directly with the customer (e.g., in a disconnected region). The customer may access each customer's respective container 1567(1)-(N) contained in a VM 1566(1)-(N) in real time. The container 1567(1)-(N) may be configured to make calls to a respective secondary VNIC 1572(1)-(N) contained in the application subnet(s) 1526 of the data plane application layer 1546, which may be contained in the container egress VCN 1568. The secondary VNIC 1572(1)-(N) may transmit the call to the NAT gateway 1538, which may transmit the call to the public Internet 1554. In this example, containers 1567(1)-(N), which may be accessed by customers in real time, may be isolated from control plane VCN 1516 and may be isolated from other entities contained in data plane VCN 1518. Containers 1567(1)-(N) may also be isolated from resources from other customers.

[0226] In other examples, a customer may use containers 1567(1)-(N) to call cloud service 1556. In this example, a customer may run code in containers 1567(1)-(N) that requests a service from cloud service 1556. Containers 1567(1)-(N) may transmit the request to secondary VNICs 1572(1)-(N), which may transmit the request to a NAT gateway, which may transmit the request to public Internet 1554. Public Internet 1554 may transmit the request to LB subnet(s) 1522 contained in control plane VCN 1516 via Internet gateway 1534. In response to determining that the request is valid, LB subnet(s) may transmit the request to application subnet(s) 1526, which may transmit the request to cloud service 1556 via service gateway 1536.

[0227] It should be appreciated that the IaaS architectures 1200, 1300, 1400, 1500 depicted in the various figures may have other components in addition to the components depicted. In addition, the embodiments shown in the various figures are merely some examples of cloud infrastructure systems that may be combined with embodiments of the present disclosure. In some other embodiments, the IaaS system may have more or fewer components than shown in the various figures, may combine two or more components, or may have a different arrangement or configuration of components.

[0228] In some embodiments, the IaaS system described herein may include application suites, middleware, and database service offerings delivered to customers in a self-service, subscription-based, elastically scalable, reliable, highly available, and secure manner. An example of such an IaaS system is the Oracle Cloud Infrastructure (OCI) provided by the present assignee.

[0229] Fig.16 An example computer system 1600 is illustrated in which various embodiments of the present disclosure may be implemented. System 1600 may be used to implement any of the computer systems described above. As shown, computer system 1600 includes a processing unit 1604 that communicates with multiple peripheral subsystems via a bus subsystem 1602. These peripheral subsystems may include a processing acceleration unit 1606, an I / O subsystem 1608, a storage subsystem 1618, and a communication subsystem 1624. Storage subsystem 1618 includes a tangible computer-readable storage medium 1622 and a system memory 1610.

[0230] Bus subsystem 1602 provides a mechanism for allowing various components and subsystems of computer system 1600 to communicate with each other by intention. Although bus subsystem 1602 is schematically shown as a single bus, alternative embodiments of bus subsystem can utilize multiple buses. Bus subsystem 1602 can be any of several types of bus structures, including a memory bus or memory controller, a peripheral bus, and a local bus using any various bus architectures. For example, this architecture can include an industrial standard architecture (ISA) bus, a microchannel architecture (MCA) bus, an enhanced ISA (EISA) bus, a video electronics standard association (VESA) local bus, and a peripheral component interconnect (PCI) bus, which can be implemented as a Mezzanine bus manufactured by the IEEE P1386.1 standard.

[0231] The processing unit 1604, which may be implemented as one or more integrated circuits (e.g., conventional microprocessors or microcontrollers), controls the operation of the computer system 1600. One or more processors may be included in the processing unit 1604. These processors may include single-core or multi-core processors. In certain embodiments, the processing unit 1604 may be implemented as one or more independent processing units 1632 and / or 1634, wherein a single-core or multi-core processor is included in each processing unit. In other embodiments, the processing unit 1604 may also be implemented as a quad-core processing unit formed by integrating two dual-core processors into a single chip.

[0232] In various embodiments, the processing unit 1604 can execute various programs in response to the program code and can maintain multiple concurrently executed programs or processes. At any given time, some or all of the program code to be executed may reside in (one or more) processors 1604 and / or in the storage subsystem 1618. Through appropriate programming, (one or more) processors 1604 can provide the various functions described above. The computer system 1600 may additionally include a processing acceleration unit 1606, which may include a digital signal processor (DSP), a special processor, and the like.

[0233] I / O subsystem 1608 may include user interface input devices and user interface output devices. User interface input devices may include a keyboard, a pointing device such as a mouse or trackball, a touch pad or touch screen incorporated into a display, a scroll wheel, a click wheel, a dial, buttons, switches, a keyboard, an audio input device with a voice command recognition system, a microphone, and other types of input devices. User interface input devices may include, for example, motion sensing and / or gesture recognition devices such as Microsoft Motion sensors that enable users to control devices such as 360 game controller. The user interface input device may also include an eye gesture recognition device, such as detecting eye activity from the user (e.g., a "wink" when taking a picture and / or making a menu selection) and translating the eye gesture to an input device (e.g., Google ) In addition, the user interface input device may include a user interface that enables the user to communicate with the voice recognition system (e.g., Navigator) interactive voice recognition sensing device.

[0234] The user interface input device may also include, but is not limited to, a three-dimensional (3D) mouse, a joystick or pointing stick, a game panel and a drawing board, and audio / video equipment, such as a speaker, a digital camera, a digital camcorder, a portable media player, a webcam, an image scanner, a fingerprint scanner, a barcode reader 3D scanner, a 3D printer, a laser rangefinder, and a sight tracking device. In addition, the user interface input device may include, for example, a medical imaging input device, such as computer tomography, magnetic resonance imaging, positron emission tomography, medical ultrasound equipment. The user interface input device may also include, for example, an audio input device, such as a MIDI keyboard, a digital musical instrument, etc.

[0235] The user interface output device may include a display subsystem, indicator lights, or a non-visual display such as an audio output device, etc. The display subsystem may be a cathode ray tube (CRT), a flat panel device such as a liquid crystal display (LCD) or a plasma display, a projection device, a touch screen, etc. In general, the use of the term "output device" is intended to include all possible types of devices and mechanisms for outputting information from the computer system 1600 to a user or other computer. For example, the user interface output device may include, but is not limited to, various display devices that visually convey text, graphics, and audio / video information, such as monitors, printers, speakers, headphones, car navigation systems, plotters, voice output devices, and modems.

[0236] Computer system 1600 may include a storage subsystem 1618 containing software elements, shown presently located within system memory 1610. System memory 1610 may store program instructions executable and loadable on processing unit 1604, as well as data generated during execution of these programs.

[0237] Depending on the configuration and type of computer system 1600, system memory 1610 may be volatile (such as random access memory (RAM)) and / or non-volatile (such as read-only memory (ROM), flash memory, etc.). RAM typically contains data and / or program modules that are immediately accessible to and / or currently being operated and executed by processing unit 1604. In some embodiments, system memory 1610 may include a variety of different types of memory, such as static random access memory (SRAM) or dynamic random access memory (DRAM). In some embodiments, a basic input / output system (BIOS) containing basic routines that help transfer information between elements within computer system 1600, such as during startup, may typically be stored in ROM. By way of example and not limitation, system memory 1610 also illustrates application programs 1612, which may include client applications, web browsers, middle-tier applications, relational database management systems (RDBMS), etc., program data 1614, and an operating system 1616. As an example, operating system 1616 may include various versions of Microsoft Apple and / or Linux operating systems, various commercially available or UNIX-like operating systems (including but not limited to various GNU / Linux operating systems, Google OS, etc.) and / or such as iOS, Phone, OS, 16OS and OS operating system's mobile operating system.

[0238] Storage subsystem 1618 may also provide a tangible computer-readable storage medium for storing basic programming and data structures that provide the functionality of some embodiments. Software (programs, code modules, instructions) that provide the above-described functionality when executed by the processor may be stored in storage subsystem 1618. These software modules or instructions may be executed by processing unit 1604. Storage subsystem 1618 may also provide a repository for storing data used in accordance with the present disclosure.

[0239] Storage subsystem 1600 may also include a computer-readable storage media reader 1620, which may be further connected to computer-readable storage media 1622. Together with system memory 1610, and optionally in conjunction therewith, computer-readable storage media 1622 may comprehensively represent remote, local, fixed, and / or removable storage devices plus storage media for temporarily and / or more permanently containing, storing, transferring, and retrieving computer-readable information.

[0240] The computer-readable storage medium 1622 containing the code or portions of the code may also include any suitable media known or used in the art, including storage media and communication media, such as, but not limited to, volatile and non-volatile, removable and non-removable media implemented in any method or technology for storage and / or transmission of information. This may include tangible computer-readable storage media, such as RAM, ROM, electrically erasable programmable ROM (EEPROM), flash memory or other memory technology, CD-ROM, digital versatile disk (DVD) or other optical storage device, magnetic cassette, magnetic tape, magnetic disk storage device or other magnetic storage device, or other tangible computer-readable medium. This may also include non-tangible computer-readable media, such as data signals, data transmissions, or any other medium that can be used to transmit the desired information and can be accessed by the computing system 1600.

[0241] By way of example, computer-readable storage media 1622 may include a hard drive that reads from or writes to non-removable nonvolatile magnetic media, a magnetic disk drive that reads from or writes to removable nonvolatile magnetic disks, and a magnetic disk drive that reads from or writes to removable nonvolatile optical disks (such as CD ROMs, DVDs, and optical drives). Computer readable storage media 1622 may include, but are not limited to: The computer readable storage medium 1622 may include a solid-state drive (SSD) based on non-volatile memory (such as a flash memory-based SSD, an enterprise flash drive, a solid-state ROM, etc.), a volatile memory-based SSD (such as a solid-state RAM, dynamic RAM, static RAM, DRAM-based SSD), a magnetoresistive RAM (MRAM) SSD, and a hybrid SSD using a combination of a DRAM-based SSD and a flash memory-based SSD. The disk drive and its associated computer-readable media may provide non-volatile storage of computer-readable instructions, data structures, program modules, and other data for the computer system 1600.

[0242] The communication subsystem 1624 provides an interface to other computer systems and networks. The communication subsystem 1624 is used as an interface for receiving data from other systems and transmitting data from the computer system 1600 to other systems. For example, the communication subsystem 1624 can enable the computer system 1600 to be connected to one or more devices via the Internet. In some embodiments, the communication subsystem 1624 may include a radio frequency (RF) transceiver component for accessing a wireless voice and / or data network (e.g., using cellular phone technology, advanced data network technology such as 3G, 4G or EDGE (Enhanced Data Rate for Global Evolution), WiFi (IEEE 802.11 series standards), or other mobile communication technologies, or any combination thereof), a global positioning system (GPS) receiver component and / or other components. In some embodiments, the communication subsystem 1624 may provide a wired network connection (e.g., Ethernet) as an addition or replacement of a wireless interface.

[0243] In some embodiments, the communication subsystem 1624 may also receive incoming communications in the form of structured and / or unstructured data feeds 1626 , event streams 1628 , event updates 1630 , and the like on behalf of one or more users who may use the computer system 1600 .

[0244] For example, the communication subsystem 1624 may be configured to receive data feeds 1626 in real time from users of social networks and / or other communication services, such as feed, Updates, web feeds such as Rich Site Summary (RSS) feeds, and / or real-time updates from one or more third-party information sources.

[0245] Additionally, the communication subsystem 1624 may also be configured to receive data in the form of continuous data streams, which may include event streams 1628 and / or event updates 1630, which may be continuous or unbounded in nature, real-time events without explicit termination. Examples of applications that generate continuous data may include, for example, sensor data applications, financial quote machines, network performance measurement tools (e.g., network monitoring and traffic management applications), clickstream analysis tools, automotive traffic monitoring, and the like.

[0246] The communication subsystem 1624 can also be configured to output structured and / or unstructured data feeds 1626, event streams 1628, event updates 1630, etc. to one or more databases, which can communicate with one or more streaming data source computers coupled to the computer system 1600.

[0247] Computer system 1600 may be one of various types, including a handheld portable device (e.g., Cellular phone, computing tablets, PDAs), wearable devices (e.g., Google head mounted display), PC, workstation, mainframe, kiosk, server rack, or any other data processing system.

[0248] Due to the ever-changing nature of computers and networks, the description of the computer system 1600 depicted in the figure is intended only as a specific example. Many other configurations with more or fewer components than the system depicted in the figure are possible. For example, customized hardware and / or specific elements can be implemented with hardware, firmware, software (including small applications (applets)) or a combination thereof. In addition, connections to other computing devices such as network input / output devices can also be used. Based on the disclosure and teachings provided herein, those of ordinary skill in the art will recognize other ways and / or methods of implementing various embodiments.

[0249] Although specific embodiments have been described, various modifications, variations, alternative constructions, and equivalents are also included within the scope of this disclosure. Embodiments are not limited to operating within certain specific data processing environments, but can be freely operated within multiple data processing environments. In addition, although embodiments have been described using a specific series of transactions and steps, it should be clear to those skilled in the art that the scope of this disclosure is not limited to the described series of transactions and steps. The various features and aspects of the above-described embodiments may be used alone or in combination.

[0250] In addition, although embodiments have been described using a specific combination of hardware and software, it should be appreciated that other combinations of hardware and software are also within the scope of the present disclosure. Embodiments may be implemented using only hardware, or only software, or using a combination thereof. The various processes described herein may be implemented in any combination on the same processor or on different processors. Thus, where a component or module is described as being configured to perform certain operations, such configuration may be accomplished by, for example, designing an electronic circuit to perform the operation, by programming a programmable electronic circuit (such as a microprocessor) to perform the operation, or by any combination thereof. Processing may communicate using a variety of techniques, including but not limited to conventional techniques for interprocess communication, and different process pairs may use different techniques, or the same process pair may use different techniques at different times.

[0251] The description and drawings are therefore to be regarded in an illustrative rather than a restrictive sense. However, additions, subtractions, deletions and other modifications and changes may be made without departing from the broader spirit and scope set forth in the claims. Therefore, while specific disclosed embodiments have been described, these are not intended to be limiting. Various modifications and equivalents are within the scope of the following claims.

[0252] The use of the terms "one" and "an" and "the" and similar references in the context of describing the disclosed embodiments (especially in the context of the following claims) should be interpreted as covering both the singular and the plural, unless otherwise indicated herein or clearly contradictory to the context. Unless otherwise stated, the terms "include", "have", "include" and "contain" should be interpreted as open terms (i.e., meaning "including but not limited to"). The term "connected" should be interpreted as partially or entirely included, attached to or connected together, even if there is something in the middle. Unless otherwise indicated herein, the description of the value range in this article is only intended to be used as a shorthand method for individually quoting each individual value falling within the range, and each individual value is incorporated into the specification as if it is described separately in this article. Unless otherwise indicated herein or clearly contradictory to the context, all methods described herein can be performed in any suitable order. Unless otherwise stated, the use of any and all examples or exemplary language (e.g., "such as") provided herein is only intended to better illustrate the embodiment and does not limit the scope of the present disclosure. Any language in the specification should not be interpreted as indicating that any unclaimed element is essential to the practice of the present disclosure.

[0253] Unless expressly stated otherwise, disjunctive language (such as the phrase "at least one of X, Y, or Z") is intended to be understood in context as serving generally to indicate that an item, term, etc. may be any one of X, Y, or Z, or any combination thereof (e.g., X, Y, and / or Z). Thus, such disjunctive language is generally not intended to, and should not, imply that certain embodiments require that at least one of X, at least one of Y, or at least one of Z each be present.

[0254] Preferred embodiments of the present disclosure are described herein, including the best mode known for implementing the present disclosure. Variations of these preferred embodiments will become apparent to those of ordinary skill in the art after reading the above description. Those of ordinary skill should be able to appropriately adopt such variations and can practice the present disclosure in a manner different from that specifically described herein. Thus, the present disclosure includes all modifications and equivalent forms of the subject matter recorded in the appended claims where permitted by applicable law. In addition, unless otherwise indicated herein, the present disclosure includes any combination of the above-mentioned elements in all possible variations thereof.

[0255] All references (including publications, patent applications and patents) cited herein are incorporated herein by reference to the same extent, as if indicating that each reference is individually and specifically incorporated by reference and set forth in full herein. In the aforementioned specification, various aspects of the present disclosure are described with reference to their specific embodiments, but it will be appreciated by those skilled in the art that the present disclosure is not limited thereto. Each feature and aspect disclosed above can be used individually or in combination. In addition, without departing from the broader spirit and scope of this specification, the embodiment can be used in any number of environments and applications other than the environment and application described herein. Thus, this specification and the accompanying drawings should be considered to be illustrative rather than restrictive.

Claims

1. A method comprising: extracting, by a network device in a communication path between a source host machine and a destination host machine, a tag from a packet received by the network device, the packet originating from a source executing on the source host machine and destined for the destination host machine, the tag being set by the source and indicating a first traffic class to be associated with the packet, the first traffic class being selected by the source from a plurality of traffic classes; determining, by the network device, a first traffic class based on the label extracted from the packet; as well as The packet is processed by the network device based on the first traffic class.

2. The method of claim 1, wherein the network device is a top-of-rack switch or a spine switch.

3. The method of claim 1 or 2, wherein the source host machine is included in a first rack of the network cluster and the destination host machine is included in a second rack of the network cluster, the second rack being different from the first rack.

4. The method of claim 1, 2 or 3, wherein the tag is included in a service type field included in a header part of the packet, the service type field being 8 bits long, and wherein 6 bits of the service type field are allocated to the tag.

5. The method of any of the preceding claims, wherein the plurality of traffic classes comprises bandwidth-sensitive RDMA traffic, latency-sensitive RDMA traffic, management-type traffic, or lossy TCP traffic.

6. The method of any preceding claim, wherein the source corresponds to an application executing on a source host machine.

7. The method according to any one of the preceding claims, further comprising: A first set of parameters associated with a network device and / or a second set of parameters corresponding to a network interface card associated with the source host machine are set based on the first traffic class.

8. The method of claim 7, wherein the first set of parameters associated with the network device includes a first parameter corresponding to a type of explicit congestion notification marking and a second parameter corresponding to a depth of a queue associated with the network device.

9. The method of claim 7 or 8, wherein the second set of parameters corresponding to the network interface card associated with the source host machine includes an adaptive retransmission parameter, a slow restart parameter, a queue number parameter, and a packet sequence number parameter.

10. A network device comprising: processor; as well as A memory comprising instructions that, when executed by a processor, cause the network device to at least perform the following operations: extracting, by the network device in a communication path between a source host machine and a destination host machine, a tag from a packet received by the network device, the packet originating from a source executing on the source host machine and destined for the destination host machine, the tag being set by the source and indicating a first traffic class to be associated with the packet, the first traffic class being selected by the source from a plurality of traffic classes; determining a first traffic class based on the label extracted from the packet; as well as The packet is processed based on the first traffic class.

11. The network device of claim 10, wherein the network device is a top-of-rack switch or a spine switch.

12. The network device of claim 10 or 11, wherein the source host machine is included in a first rack of the network cluster and the destination host machine is included in a second rack of the network cluster, the second rack being different from the first rack.

13. The network device of claim 10, 11 or 12, wherein the tag is included in a service type field included in a header part of the packet, the service type field being 8 bits long, and wherein 6 bits of the service type field are allocated to the tag.

14. The network device of any one of claims 10 to 13, wherein the plurality of traffic classes comprises bandwidth-sensitive RDMA traffic, latency-sensitive RDMA traffic, management-type traffic, or lossy TCP traffic.

15. The network device of any one of claims 10 to 14, wherein the source corresponds to an application executing on a source host machine.

16. The network device according to any one of claims 10 to 15, further configured to: A first set of parameters associated with a network device and / or a second set of parameters corresponding to a network interface card associated with the source host machine are set based on the first traffic class.

17. The network device of claim 16, wherein the first set of parameters associated with the network device includes a first parameter corresponding to a type of explicit congestion notification marking and a second parameter corresponding to a depth of a queue associated with the network device.

18. The network device of claim 16 or 17, wherein the second set of parameters corresponding to the network interface card associated with the source host machine includes an adaptive retransmission parameter, a slow restart parameter, a queue number parameter, and a packet sequence number parameter.

19. A non-transitory computer-readable medium storing specific computer-executable instructions that, when executed by a processor, cause a computer system to at least perform the following operations: extracting, by a network device in a communication path between a source host machine and a destination host machine, a tag from a packet received by the network device, the packet originating from a source executing on the source host machine and destined for the destination host machine, the tag being set by the source and indicating a first traffic class to be associated with the packet, the first traffic class being selected by the source from a plurality of traffic classes; determining, by the network device, a first traffic class based on the label extracted from the packet; as well as The packet is processed by the network device based on the first traffic class.

20. The non-transitory computer readable medium storing the specific computer executable instructions according to claim 19, wherein the computer system is further configured to: A first set of parameters associated with a network device and / or a second set of parameters corresponding to a network interface card associated with the source host machine are set based on the first traffic class.

Images