Packet drop detection in the local networking stack via packet correlation

By instantiating packet monitoring applications in the virtualized network stack in the data center, tracking and reporting packet propagation and discarding information, the problem of identification of latency problems in the virtual network stack is solved, and rapid diagnosis and improvement of network speed is achieved.

CN113647060BActive Publication Date: 2024-11-01MICROSOFT TECHNOLOGY LICENSING LLC
View PDF 2 Cites 0 Cited by

Patent Information

Application Number
CN202080026932.7
Authority / Receiving Office
CN · China
Patent Type
Patents(China)
Current Assignee / Owner
Priority Date
2019-05-31
Filing Date
2020-03-25
Publication Date
2024-11-01
Estimated Expiration
2040-03-25

AI Technical Summary

Technical Problem

In data centers, it is difficult for customers to determine the root cause of latency in the virtual networking stack when accessing virtual machines or containers, making latency issues difficult to quickly identify and resolve.

Method used

Instantiate the packet monitoring application on the server hosting the virtualized network stack, tracks packet propagation and discarding of each component through the packet monitoring application, provides an API for component reporting packet propagation and discarding information, and allows users to enter filter parameters to select tracking Or ignore specific packets and use association IDs to implement packet tracking and delay calculation across components.

Benefits of technology

It significantly reduces the time to identify the root cause of delay problems, improves the efficiency of diagnosis of network problems, saves network bandwidth, and improves network speed on customer devices.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN113647060B_ABST
    Figure CN113647060B_ABST
Patent Text Reader

Abstract

A packet monitoring application instantiated on a server hosting the virtualized network stack is used to track packet propagation and drops at each component within the network stack to reduce the amount of time to identify the root cause of latency issues. Administrators can selectively enable or disable the packet monitoring application. Components within the virtualized network stack report packet drops and successful packet propagation to the packet monitoring application, which can filter packets based on input parameters. Thus, a user can select the level of granularity at which packets are filtered in the virtualized network stack while being able to evaluate the traversal of each packet through each component within the network stack. The packet monitoring application can also perform post-processing on the filtered packets to determine the latency between components or portions of the virtualized network stack.
Need to check novelty before this filing date? Find Prior Art

Description

Background Art

[0001] Data centers may set up virtual machines or containers that run applications and software provided to customers for their use. The virtualized network stack running within individual computer servers in the data center may have multiple components for receiving and delivering packets. Each component in the virtualized network stack provides unique functionality that can become a point of failure (e.g., dropped packets or increased latency) for data reception and transmission. In some cases, problems that customers discover when accessing their virtual machines or containers may make it difficult to identify the root cause of the latency occurring in the virtual networking stack. Summary of the invention

[0002] A packet monitoring application instantiated on a server hosting a virtualized network stack is used to track packet propagation and discards of each component within the network stack to reduce the amount of time to identify the root cause of a latency problem. A computer server administrator can selectively enable or disable the packet monitoring application. When disabled, components in the virtualized network stack do not report packet propagation or discards to the packet monitoring application. When enabled, the component calls an application programming interface (API) associated with the packet monitoring application and reports packet discards and propagation to the packet monitoring application. Users can enter parameters into the packet monitoring application, through which certain packets are selected to be disclosed to the user, while other packets are ignored or discarded. Filtering parameters can include custom granularity levels in virtualized network stack components, including monitoring levels within components (e.g., miniports or specific components within the virtualized network stack) and communication protocols (e.g., transmission control protocol (TCP), user datagram protocol (UDP), etc.), as well as other parameters. A unique correlation identifier (ID) is associated with each packet to enable tracking of packets across each component in the network stack. The ability to identify packet propagation and drops for each component within the networking stack enables the user to identify where within the network stack packet drops or delays occur, thereby reducing the time to identify the root cause. Post-processing of filtered packets may also be implemented, where various determinations or calculations are performed on the packets, such as packet delay at the entire or portion of the network stack, packet propagation averages through components in the network stack, and the like.

[0003] This summary is provided to introduce a series of concepts in a simplified form, which will be further described in the following detailed description. This summary is not intended to identify the key features or essential features of the claimed subject matter, nor is it intended to be used as an aid in determining the scope of the claimed subject matter. In addition, the claimed subject matter is not limited to implementations that address any or all of the shortcomings noted in any part of this disclosure. It should be understood that the above-mentioned subject matter can be implemented as a computer-controlled device, a computer process, a computing system, or as an article such as one or more computer-readable storage media. These and various other features will be apparent by reading the following detailed description and reviewing the associated drawings. BRIEF DESCRIPTION OF THE DRAWINGS

[0004] Figure 1 An exemplary environment is shown in which a remote computing device communicates packets with a computer server operating in a data center;

[0005] Figure 2 A schematic diagram showing components within a virtualized network stack;

[0006] Figure 3 A schematic diagram showing components within a virtualized network stack;

[0007] Figure 4 An illustrative environment is shown in which a user input enables a packet monitoring application;

[0008] Figure 5 A schematic diagram is shown in which components within a networking stack send reports on packet propagation and packet drops to a packet monitor application programming interface (API);

[0009] Figure 6 An illustrative environment is shown in which a computing device receives filtering parameters from a user input;

[0010] Figure 7 shows a schematic classification of parameters by which data packets are tracked and filtered in a virtualized network stack;

[0011] Figure 8 shows a schematic diagram in which packets are assigned unique associated identifiers (IDs) by a packet monitoring application;

[0012] Fig. 9 shows a schematic diagram in which a packet at each component in a virtualized network stack is followed by a unique correlation ID;

[0013] Fig.10 shows a schematic diagram in which a packet monitoring application utilizes a correlation ID to determine packet delay at a component;

[0014] Fig.11shows a schematic classification of information reported by components to a packet monitoring application;

[0015] Figure 12-14 is a flow chart of an illustrative method performed by one or more computer servers within a data center;

[0016] Fig.15 is a block diagram of an illustrative data center that may be used, at least in part, to implement current packet drop detection in a local networking stack through packet correlation;

[0017] Fig.16 is a simplified block diagram of a schematic architecture of a computing device that may be used, at least in part, to implement the present packet drop detection in a local networking stack through packet correlation; and

[0018] Fig.17 is a simplified block diagram of an illustrative remote computing device, remote service, or computer system that may be used, in part, to implement the present packet drop detection in a local networking stack through packet correlation.

[0019] Like reference numerals in the drawings represent like elements. Unless otherwise indicated, the elements are not drawn to scale. DETAILED DESCRIPTION

[0020] Figure 1 An illustrative environment is shown in which a client computing device 105 communicates packets 115 with a computer server 110 that may be operating in a data center via a network 130. The network may include a local area network, a wide area network, and the Internet. A computing device 120 local to the computer server may be configured with a user interface to enable a user 125 to control various aspects and operations on the computer server. The computer server in the data center may be associated with a cloud service provider that provides solutions to customers operating client computing devices. For example, a computer server may host virtual machines or containers, each of which is associated with and accessible to a unique customer. Virtual machines and containers may provide solutions for client computing devices, such as data analysis, data storage, artificial intelligence processing, and the like. A single computer server may host multiple virtual machines and / or containers that may be accessed by multiple different customers, respectively.

[0021] The client computing device 105 may be any number of devices configured with a network connection to connect with and utilize solutions provided by the computer server 110. Non-exhaustive types of computing devices include laptops, smart phones, tablet computers, local servers, and Internet of Things (IoT) devices (shown generally and schematically as system-on-chip (SoC) devices) that can collect and report telemetry data. A plurality of packets 115 may be transmitted between the computer server and the client computing device.

[0022] Figure 2 A schematic diagram of a virtualized network stack 200 configuration of a host computer server is shown. Each component within the networking stack can communicate with a packet monitoring application programming interface (API) 205. For example, each stack component can execute an API call 210 and report a packet 215 to a packet monitoring application 220, such as the status of the packet (e.g., packet discard and successful packet propagation). The packet 115 can be in either direction within the networking stack, i.e., delivered or received. The virtualized network stack can be used to deliver the packet to a virtual machine or container of a corresponding customer. Thus, multiple virtual machines and containers operated by multiple different customers can be operated within a data center and processed by a single computer server or multiple computer servers together.

[0023] A brief description of the components within the virtualized network stack 200 follows. Although some names utilized throughout the description and figures may be specific to The described functions may be applicable to any implementation. For example, in other implementations of virtualized network stacks, some of the described components may not be used or additional components may be used.

[0024] The networking application 225 may be used to provide control of the components to the user. The packet monitoring application 220 is presented to the user on a user interface associated with the computer server for enabling or disabling calls to the packet monitoring API 205 by components within the network stack.

[0025] A network interface controller (NIC) 230 may be a hardware device implemented to enable Wi-Fi, Ethernet, or other connections. NIC 230 may provide physical and data link layer network communications to another device such as a router. NDIS (Network Driver Interface Specification) 235 provides an API for the NIC that includes a library of functions used as a standard interface. WFP ( The Network Filtering Platform (WFP MAC 240) is a set of system services and APIs that provide a platform for creating network filtering applications. WFP allows developers to write code that interacts with the packet processing that occurs at multiple layers in the operating system's networking stack. For example, developers can implement firewalls, intrusion detection systems, antivirus programs, network monitoring tools, and parental controls before data reaches its destination. In this example, WFP filtering can be applied at the MAC address level or the IP address level (see WFP IP 270).

[0026] The virtual filtering platform (VFP) 250 is an extension of the virtual switch (vSwitch) 245 of the virtual networking stack that enables core software defined networking (SDN) functions for network services. The VFP can be a programmable network component that exposes an easily programmable abstract interface to network agents that represent network controllers such as virtual network controllers and software load balancer controllers. The SDN data plane improves the scalability of data delivery and reception by leveraging host components and performing packet processing on host computer servers running in a data center.

[0027] Pacer260 is an abbreviation term for Packet Scheduler, which is a filter intermediate driver that is provided and enforces Quality of Service (QoS) parameters for network data flows. Pacer can perform various functions to ensure QoS, including allowing administrators to prioritize network traffic and limit bandwidth utilization so that the operating system continues to perform certain operations, among other functions.

[0028] TCP / IP (Transmission Control Protocol / Internet Protocol) 265 is a communication protocol used at the transport layer and network layer of the TCP / IP stack. These layers establish reliable connections between host and client devices (such as computer servers and client computing devices) and route packets between networks. WinNAT ( Network Address Translation (NAT) 275 is used to translate private IP addresses assigned to virtual services (e.g., virtual machines, containers, or other services) running on a computer server and share public IP addresses routable to the Internet. WFP IP 270 is used as described above with respect to WFP MAC 240.

[0029] Winsock 280 is an API that handles input / output requests for Internet applications within the operating system. Winsock can run between applications (such as web browsers) and the TCP / IP layer and provide interfaces for different versions of operating systems. AFD (auxiliary function driver) 285 is a kernel driver that can be used with Winsock to enable a computer server to communicate with hardware or connected devices. WSK (Winsock kernel 1) 290 is a kernel mode network programming interface (NPI) that creates a network interface between network modules that can be attached to each other. For example, a client module registered as a client to a specific NPI can be attached to a provider module registered as a provider of the same NPI.

[0030] The HTTP (Hypertext Transfer Protocol) 295 and SMB (Server Message Block) 298 protocols are exemplary communication protocols that operate at the application layer of the TCP / IP model. HTTP utilizes requests and responses to communicate between a client and a server. SMB is a protocol that allows systems running on the same network to share files. Other application layer protocols not shown may also be implemented, including SMTP (Simple Mail Transfer Protocol) and FTP (File Transfer Protocol).

[0031] The above about Figure 2 The description provided is applicable to Figure 3 The virtualized network stack depicted in . For example Figure 3 Depicted are general terms and functionality within each networking stack 300. Propagation points 360 depict the transfer of packets from one component to the next.

[0032] The filter 305 and extension 310 associated with NDIS 235 may, for example, implement the functionality described above with respect to WFP and VFP and perform filtering functions on packets. A port NIC (ptNIC) 315 may be a port that enters or leaves a vSwitch 245, and an extension NIC (extNIC) 320 may be an interface to the extension 310 associated with NDIS 235. The ptNIC may provide an external virtual port for connectivity to an external physical network, and may provide a connection to a physical function or a virtual function. An extension protocol (extProt) 365 may be used to forward packets to the next layer within the virtualized network stack, such as a host network card (hNIC) 325 or a virtual machine network card (vmNIC) 330. The vmNIC may transmit packets to or from a virtual machine or container 335, and may be used to connect a virtual machine to the Internet or bridge to a different network segment.

[0033] The hNIC 325 may be used to transmit packets to or from operations or applications on a computer server that is distinct from a virtual machine or container. As packets pass through the hNIC, NDIS 235 and filters 305 are applied to the packets before forwarding them to the TCP / IP stack, which exemplarily shows transport 340, network 345, and frame 350 layers. Reference 355 may be a collection of functions exposed by the driver for specialized filtering and manipulation of packets, such as virus scanning, parental controls for inappropriate content, and packet data parsing.

[0034] Virtual machines / containers 335 serve different functions and can be configured differently by the customer. A virtual machine is an emulation of a computer system instantiated on a host computer server, where multiple independently running virtual machines can be used. A virtual machine includes its own virtual hardware, including a processing unit, memory, network interfaces, etc. A virtual machine can run different client operating systems and applications depending on the customer's configuration. A container can also execute processes for a customer, but may not use a client operating system like a virtual machine, but rather a container consists of application code to perform functions. A container can be isolated from the application code and the content necessary to perform the intended function of the container.

[0035] Figure 4 An illustrative environment is shown in which a user 125 enters input into a computing device 120 to enable a packet monitoring application. When the packet monitoring application is enabled, a component 405 within a virtualized network stack 410 calls a packet monitor API 205. The virtualized network stack 410 is used to represent, for example, Figure 2 or the networking stack depicted in 3. Likewise, component 405 is used to represent Figure 2 or Figure 3 Any one or more of the physical or virtual network components of the networking stack depicted in . Thus, a reference to a component may reference any one or more of the network components that make up the virtualized network stack, including NDIS, vSwitch, NIC, TCP / IP, vmNIC, filters, extensions, etc.

[0036] Figure 5 2 shows a schematic diagram in which a registered component 405 sends a report 505 to a packet monitor API 205. The report includes packet information for successful packet propagation 510 and packet drops 515 at the component. Successful packet propagation can be evaluated and reported based on the boundaries (or edges) of the component, such as where packets enter and leave the component (see, e.g., Figure 3 The propagation point 360 in the component). Packet drops can be evaluated and reported at any point within the component so that the packet monitor API has information about where the packet drops occurred within the component. This information can be used to determine whether a specific function within the component is interfering with packet propagation.

[0037] Figure 6An illustrative environment is shown in which a user 125 uses a computing device 120 to input packet filtering parameters used by a packet monitoring application 220. The packet monitoring application uses these parameters to evaluate which packet information to select for display on a user interface on the computing device 120 and which packet information to ignore or disregard. Depending on the scenario, certain packets may be more demonstrative than other types of packets and related to latency issues experienced by a customer. Connectivity issues experienced by a customer may indicate to an administrator of a computer server where a virtualized network stack resides that Hypertext Transfer Protocol (HTTP) packets are related to identifying where on the network stack latency or packet drops are occurring. Server Message Block (SMB) or other protocols may not be relevant to a user's browser connectivity issues and are therefore filtered out and not presented to the user by the packet monitoring application 220.

[0038] Figure 7 705, a schematic taxonomy of parameters by which a packet monitoring application filters received packets within a virtualized network stack. Non-exhaustive parameters include monitoring level (e.g., only miniports, specific components, all components, etc.) 710, media access control (MAC) source and / or destination 715, Ethernet protocol (e.g., any, IPv4 (Internet Protocol version 4), IPv6, ICMP (Internet Control Message Protocol), ICMPv6) 720, VLAN (Virtual Local Area Network) ID 725, IP address source or destination (e.g., any one or both addresses) 730, communication protocol (e.g., any TCP (Transmission Control Protocol), UDP (User Datagram Protocol)) 735, TCP flags (e.g., TCP SYN (Synchronize), TCP RST (Reset) or TCP FIN (Complete)) 740, port source and / or destination (e.g., any one or both ports) 745, packet encapsulation (inspect inner and / or outer packets) 750, and service type (e.g., storage, network-based, artificial intelligence, etc.) 755. Any one or more of these parameters may be entered and configured into a packet monitoring application to filter and track specific information packet configurations that may help troubleshoot packet drop or delay issues experienced by a customer.

[0039] Figure 8 Schematic diagram is shown in which the packet monitoring application 220 assigns 805 an association identifier (ID) 810 to each packet 115. The association ID is unique to a particular packet and is used throughout each component 405 ( Figure 4). The correlation ID can be, for example, a 32-bit number on a 64-bit operating system or a 16-bit number on a 32-bit operating system. The deployment of the correlation ID enables packet monitoring applications to display the data path of the packet, which can provide information for detecting packet delays, packet drops, etc. Thus, the user can utilize parameters to filter the packets to be tracked or dropped, and then view each individual packet or a group of similar packets that traverse the virtualized network stack.

[0040] Fig. 9 A schematic diagram is shown where a correlation ID 810 follows the packet 115 through each component 405 that the packet passes through. Each component may be a different network component in a virtualized network stack (e.g., a vSwitch, a filter, a virtual port, etc.). The packet monitoring application 220 ( Figure 2 ) is configured to display to a user each component through which a single packet traverses using its assigned correlation ID. The correlation ID is preserved for each packet through packet transformations, such as packet encapsulation, network address translation, etc. The correlation ID is associated with the original packet, so any transformations to the packet are identifiable, so the original packet can still be identified. That is, for example, the outer header on an inner packet is identifiable, so the inner packet can still be tracked throughout the virtualized network stack.

[0041] Fig.10 A schematic diagram is shown in which the packet monitoring application 220 can perform post-processing 1005 on the filtered packets using the correlation ID 810 associated with the corresponding packet 115. The post-processing tasks can include the packet monitoring application evaluating the scenario in which the customer problem occurred 1010, determining the packet delay at the component 1015, determining the duration 1020 that the packet took to traverse a portion of the networking stack, and calculating the average value of the packet's propagation across a defined portion of the network stack or the entire network stack.

[0042] Regarding the evaluation of the scenario, the packet monitoring application may trigger certain points in the virtualized network stack, across one or more computer servers that the packet passes through. For example, in a scenario where load balancing is performed for customer-related issues, the packet monitoring application can be enabled on the computer server that performs load balancing, the receiving server to which the packet is directed, and the originating server from which the packet may be sent to the customer's computing device. If the data is not load balanced, the packet monitoring application can be enabled on the computer server being used. The association ID can be used across computer servers to achieve consistent review of the packet in its path through. Other scenarios are also possible. The scenario can be implemented by user input at a user interface associated with the computer server, or it can be an automated process.

[0043] The components may report the timestamps of each data packet to the packet monitoring application. The timestamps may be associated with the boundaries of the components (e.g., ingress and egress points) or within the corresponding components (e.g., for packet drops). The ability to track packets that pass through the components enables the packet monitoring application to perform at least some post-processing 1005 operations using the associated timestamps. For example, subtracting the timestamp value of a packet when it is in one component from another timestamp value in a subsequent component may provide the latency of the packet, at least for those components. This information may also be used to calculate ergodic averages and other calculations.

[0044] The ability to track packets at a customizable level of granularity by virtualizing each component of the network stack enables users to quickly identify the root cause of latency issues that users are experiencing. This can save network bandwidth by resolving packet drops and latency issues that might otherwise prevent packets from propagating. Automated processing that can identify increased latency relative to expected latency at a component or collection of components can further speed up root cause determination to enable users to identify and resolve issues. The result can be higher network speeds on computer servers, virtual machines and containers running on computer servers, and on customer devices that utilize services provided by computer servers and data centers.

[0045] Fig.11 A classification of information 1105 of data packets 115 reported by a component to a packet monitoring application 220 is shown. The reported information may include a component name 1110, a drop function name and line number (for drop) 1115, a drop reason 1120 from a global ENUM (enumeration) list (for drop), packet payload or header information 1125 including payload type (e.g., Ethernet name, IP frame, etc.) 1130, TCP / UDP port (e.g., source and / or destination) 1135, packet direction (e.g., send or receive) 1140, user mode process name 1145, interface ID 1150, instance ID (e.g., NDIS handle) 1155, network buffer list (NBL) out-of-band (OOB) information 1160, Ethernet header information (e.g., source, destination, protocol) 1165, VLAN ID 1170, and IP header information (e.g., source, destination, protocol) 1175.

[0046] Figure 12-14 1 is a flowchart of illustrative methods 1200, 1300, and 1400 that may be performed using computer server 110 or computing device 120. Unless otherwise specified, the flowcharts and the accompanying text are not limited to a particular order or sequence. In addition, some methods or steps thereof may occur or be performed simultaneously, and depending on the requirements of the implementation, not all methods or steps must be performed in a given implementation, and some methods or steps may be optionally utilized.

[0047] In step 1205, Fig.12 In step 1210, the computer server enables operation of the packet monitoring application, which triggers each component in the virtualized network stack to call an API associated with the packet monitoring application. In step 1215, the computer server receives packets from components within the virtualized network stack at the packet monitoring application. In step 1220, the computer server uses the parameters through the packet monitoring application to filter the packets received from the components. In step 1225, the computer server assigns a correlation ID to each packet at the packet monitoring application to enable correlation and tracking of the packets as the packets pass through the components of the virtualized network stack.

[0048] In step 1305, Fig.13 In step 1310, the computer server communicates packets with one or more client computing devices, wherein during the communication, the packets pass through components of a virtualized network stack on the computer server. In step 1315, the components of the virtualized network stack report the traversal status of the packets to a packet monitoring application. In step 1320, the computer server uses the association ID to track packets across components on the virtualized network stack. In step 1325, the computer server configures the packet monitoring application to filter the packets according to the set parameters, thereby selecting corresponding packets to be displayed on a user interface associated with the computer server.

[0049] In step 1405, Fig.14 In step 1410, the computer server displays a user interface to enable a packet monitoring application, which enables a component in a network stack to report packets to the packet monitoring application. In step 1410, the computer server receives a selection of one or more parameters at a user interface, which details which packets the packet monitoring application is to select. In step 1415, a component in the network stack calls an API associated with the packet monitoring application. In step 1420, the packet monitoring application receives packets from a component of the network stack. In step 1425, the packet monitoring application uses parameters to filter received packets, wherein packets that meet the parameters are selected by the packet monitoring application to be displayed on the user interface. In step 1430, the packet monitoring application assigns a unique association ID to the filtered packets, wherein the association ID assigned to the packet follows the corresponding packet through the network stack. In step 1435, the packet monitoring application uses information about packet propagation to calculate the delay of the packet at the component. In step 1440, the computer server displays a report including the calculated delay of the filtered packets and the packet discard.

[0050] Fig.151 is a high-level block diagram of an exemplary data center 1500 that provides cloud computing services or distributed computing services that can be used to implement current packet drop detection in a local network stack through packet correlation. Multiple servers 1501 are managed by a data center management controller 1502. Load balancer 1503 distributes requests and computing workloads across servers 1501 to avoid situations in which a single server may become overwhelmed. Load balancer 1503 maximizes the available capacity and performance of resources in data center 1500. Routers / switches 1504 support data traffic between servers 1501 and between data center 1500 and external resources and users (not shown) via an external network 1505, which can be, for example, a local area network (LAN) or the Internet.

[0051] The servers 1501 may be stand-alone computing devices, and / or they may be configured as individual blades in a rack of one or more server devices. The servers 1501 have input / output (I / O) connectors 1506 that manage communications with other database entities. One or more host processors 1507 on each server 1501 run a host operating system (O / S) 1508 that supports multiple virtual machines (VMs) 1509. Each VM 1509 may run its own O / S, such that each VM O / S 1510 on a server is different, or the same, or both. The VM O / S 1510 may be, for example, different versions of the same O / S (e.g., different VMs running different current and older versions of the Windows? operating system). Additionally or alternatively, the VM O / S 1510 may be provided by different manufacturers (e.g., some VMs running different current and older versions of the Windows? operating system). operating system, while other VMs run Each VM 1509 may also run one or more applications (Apps) 1511. Each server 1501 also includes a storage device 1512 (e.g., a hard disk drive (HDD)) and a memory 1513 (e.g., RAM) that can be accessed and used by the host processor 1507 and the VM 1509 for storing software code, data, etc. In one embodiment, the VM 1509 may use the data plane API disclosed herein.

[0052] Data center 1500 provides pooled resources, and customers can dynamically provision and scale applications as needed without adding servers or additional networking. This allows customers to obtain the computing resources they need without having to procure, provision, and manage infrastructure on a per-application basis. Cloud computing data center 1500 allows customers to dynamically scale resources up or down to meet the current needs of their business. In addition, data center operators can provide usage-based services to customers so that they only pay for the resources they use when they need them. For example, a customer may initially use one VM 1509 on server 15011 to run their application 1511. When the demand for application 1511 increases, data center 1500 can add more VMs on the same server 15011 and / or new servers 1501 as needed. N Additional VMs 1509 are activated at a later time. If the demand for the application decreases later, these additional VMs 1509 can be deactivated.

[0053] The data center 1500 can provide guaranteed availability, disaster recovery, and backup services. For example, the data center can designate a VM 1509 on a server 15011 as the primary location for a customer application, and can activate a second VM 1509 on the same or different server as a standby or backup in case the first VM or server 15011 fails. The data center management controller 1502 automatically transfers incoming user requests from the primary VM to the backup VM without customer intervention. Although the data center 1500 is shown as a single location, it should be understood that the servers 1501 can be distributed to multiple locations around the world to provide additional redundancy and disaster recovery capabilities. In addition, the data center 1500 can be a local private system that provides services to a single enterprise user, or it can be a publicly accessible distributed system that provides services to multiple unrelated customers, or it can be a combination of both.

[0054] Domain Name System (DNS) server 1514 resolves domain names and host names to IP (Internet Protocol) addresses for all roles, applications, and services in data center 1500. DNS log 1515 maintains a record of which domain names have been resolved by roles. It should be understood that DNS is used as an example herein, and other name resolution services and domain name record services may be used to identify dependencies.

[0055] Datacenter health monitoring 1516 monitors the health of the physical systems, software, and environment in datacenter 1500. Health monitoring 1516 provides feedback to datacenter managers when problems are detected with servers, blades, processors, or applications in datacenter 1500, or when the network experiences bandwidth or communication issues.

[0056] Fig.16An illustrative architecture 1600 is shown for a client computing device, such as a laptop or personal computer, for packet drop detection in a local network stack through packet correlation. Fig.16 The architecture 1600 shown in FIG. 1 includes one or more processors 1602 (e.g., central processing units, dedicated artificial intelligence chips, graphics processing units, etc.); system memory 1604, including RAM (random access memory) 1606 and ROM (read only memory) 1608; and a system bus 1610 that operatively and functionally couples the components in the architecture 1600. A basic input / output system containing basic routines that help transfer information between elements within the architecture 1600, such as during startup, is typically stored in ROM 1608. The architecture 1600 also includes a mass storage device 1612 for storing software code or other computer executable code used to implement applications, file systems, and operating systems. The mass storage device 1612 is connected to the processor 1602 through a mass storage controller (not shown) connected to the bus 1610. The mass storage device 1612 and its associated computer-readable storage media provide non-volatile storage for the architecture 1600. Although the descriptions of computer-readable storage media contained herein refer to mass storage devices, such as a hard disk or a CD-ROM drive, those skilled in the art will appreciate that computer-readable storage media may be any available storage media that can be accessed by architecture 1600 .

[0057] By way of example and not limitation, computer-readable storage media may include volatile and nonvolatile, removable and non-removable media implemented in any method or technology for storage of information such as computer-readable instructions, data structures, program modules or other data. For example, computer-readable media include, but are not limited to, RAM, ROM, EPROM (erasable programmable read-only memory), EEPROM (electrically erasable programmable read-only memory), flash memory or other solid-state storage technology, CD-ROM, DVD, HD-DVD (high-definition DVD), Blu-ray or other optical storage, cassettes, magnetic tape, disk storage or other magnetic storage devices, or any other medium that can be used to store the desired information and can be accessed by the architecture 1600.

[0058] According to various embodiments, the architecture 1600 can operate in a networked environment using logical connections to remote computers through a network. The architecture 1600 can be connected to the network via a network interface unit 1616 connected to the bus 1610. It will be appreciated that the network interface unit 1616 can also be used to connect to other types of networks and remote computer systems. The architecture 1600 can also include an input / output controller 1618 for receiving and processing input from a plurality of other devices, including a keyboard, a mouse, a touch pad, a touch screen, a control device such as buttons and switches, or an electronic stylus ( Fig.16Similarly, input / output controller 1618 may provide signals to a display screen, user interface, printer, or other type of output device (also not shown). Fig.16 ) provides output.

[0059] It is understood that the software components described herein can convert the processor 1602 and the entire architecture 1600 from a general-purpose computing system to a customized special-purpose computing system when loaded into the processor 1602 and executed to facilitate the functions described herein. The processor 1602 can be composed of any number of transistors or other discrete circuit elements, which can present any number of states individually or collectively. More specifically, the processor 1602 can operate as a finite state machine in response to the executable instructions contained in the software modules disclosed herein. These computer executable instructions can convert the processor 1602 by specifying how the processor 1602 converts between states to convert the transistors or other discrete hardware elements that constitute the processor 1602.

[0060] Encoding the software modules presented herein can also transform the physical structure of the computer-readable storage medium presented herein. The specific transformation of the physical structure may depend on various factors in different implementations of this specification. Examples of these factors may include, but are not limited to, the technology used to implement the computer-readable storage medium, whether the computer-readable storage medium is characterized as primary storage or auxiliary storage, etc. For example, if the computer-readable storage medium is implemented as a semiconductor-based memory, the software disclosed herein can be encoded on the computer-readable storage medium by converting the physical state of the semiconductor memory. For example, the software can convert the state of the transistors, capacitors, or other discrete circuit elements that constitute the semiconductor memory. The software can also convert the physical state of these components to facilitate storing data thereon.

[0061] As another example, the computer readable storage medium disclosed herein can be realized using magnetic or optical technology. In such an embodiment, when the software is encoded therein, the software presented herein can convert the physical state of magnetic or optical media. These conversions can include changing the magnetic properties of a specific location in a given magnetic medium. These conversions can also include changing the physical characteristics or characteristics of a specific location in a given optical medium to change the optical properties of those locations. Without departing from the scope and spirit of this description, other conversions of physical media are possible, and the above examples are provided only for the convenience of discussion.

[0062] The architecture 1600 may also include one or more sensors 1614 or a battery or power source 1620. Sensors may be coupled to the architecture to acquire data about the environment or components, including temperature, pressure, etc. Exemplary sensors may include thermometers, accelerometers, smoke or gas sensors, pressure sensors (barometric or physical), light sensors, ultrasonic sensors, gyroscopes, etc. The power source may employ an AC power cord or a battery, such as a rechargeable battery for portability.

[0063] In view of the above, it can be appreciated that many types of physical transformations occur in architecture 1600 in order to store and execute the software components presented herein. It can also be appreciated that architecture 1600 can include other types of computing devices, including wearable devices, handheld computers, embedded computer systems, smart phones, PDAs, and other types of computing devices known to those skilled in the art. It is also contemplated that architecture 1600 may not include Fig.16 All components shown in the Fig.16 Other components not explicitly shown in the Fig.16 A completely different architecture than the one shown in .

[0064] Fig.1717 is a simplified block diagram of an illustrative computer system 1700, such as a PC or server, with which the current packet drop detection in the local network stack can be implemented through packet correlation. The computer system 1700 includes a processor 1705, a system memory 1711, and a system bus 1714, which couples various system components including the system memory 1711 to the processor 1705. The system bus 1714 can be any of several types of bus structures including a memory bus or memory controller, a peripheral bus, or a local bus using any of a variety of bus architectures. The system memory 1711 includes a read-only memory (ROM) 1717 and a random access memory (RAM) 1721. A basic input / output system (BIOS) 1725, containing basic routines such as those that help transfer information between elements within the computer system 1700 during startup, is stored in the ROM 1717. The computer system 1700 may also include a hard disk drive 1728 for reading and writing an internally disposed hard disk (not shown), a disk drive 1730 for reading or writing a removable disk 1733 (e.g., a floppy disk), and an optical drive 1738 for reading or writing a removable optical disk 1743, such as a CD (compact disk), a DVD (digital versatile disk), or other optical media. The hard disk drive 1728, the disk drive 1730, and the optical drive 1738 are connected to the system bus 1714 via a hard disk drive interface 1746, a disk drive interface 1749, and an optical drive interface 1752, respectively. The drives and their associated computer-readable storage media provide non-volatile storage of computer-readable instructions, data structures, program modules, and other data for the computer system 1700. Although the illustrative example includes a hard disk, a removable disk 1733, and a removable optical disk 1743, other types of computer-readable storage media that can store data accessible by a computer, such as magnetic tapes, flash memory cards, digital video disks, data cartridges, random access memory (RAM), read-only memory (ROM), etc., may also be used in some applications of the current packet drop detection by packet correlation in the local network stack. In addition, as used herein, the term computer-readable storage medium includes one or more instances of a type of medium (e.g., one or more disks, one or more CDs, etc.). For purposes of this specification and claims, the phrase "computer-readable storage medium" and its variations are intended to cover non-transitory embodiments and do not include waves, signals, and / or other transient and / or intangible communication media.

[0065] A plurality of program modules may be stored on a hard disk, disk 1733, optical disk 1743, ROM 1717, or RAM 1721, including an operating system 1755, one or more applications 1757, other program modules 1760, and program data 1763. A user may enter commands and information into the computer system 1700 through input devices such as a keyboard 1766 and a pointing device 1768 such as a mouse. Other input devices (not shown) may include a microphone, a joystick, a game controller, a satellite dish, a scanner, a trackball, a touch pad, a touch screen, a touch-sensitive device, a voice command module or device, a user motion or user gesture capture device, and the like. These and other input devices are typically connected to the processor 1705 through a serial port interface 1771, which is coupled to the system bus 1714, but may also be connected through other interfaces, such as a parallel port, a game port, or a universal serial bus (USB). A monitor 1773 or other type of display device is also connected to the system bus 1714 via an interface, such as a video adapter 1775. In addition to the monitor 1773, personal computers typically include other peripheral output devices (not shown) such as speakers and printers. Fig.17 The illustrative example shown in FIG. 1 also includes a host adapter 1778 , a small computer system interface (SCSI) bus 1783 , and an external storage device 1776 connected to the SCSI bus 1783 .

[0066] The computer system 1700 may use logical connections to one or more remote computers in a networked environment, such as a remote computer 1788. The remote computer 1788 may be selected as another personal computer, a server, a router, a network PC, a peer device or other public network node, and typically includes many or all of the elements described above with respect to the computer system 1700, although Fig.17 Only a single representative remote memory / storage device 1790 is shown. Logical Connections Fig.17 Depicted in the diagram are a local area network (LAN) 1793 and a wide area network (WAN) 1795. Such networking environments are often deployed in, for example, offices, enterprise-wide computer networks, intranets, and the Internet.

[0067] When used in a LAN networking environment, the computer system 1700 is connected to the local area network 1793 through a network interface or adapter 1796. When used in a WAN networking environment, the computer system 1700 typically includes a broadband modem 1798, a network gateway, or other means for establishing communications over the wide area network 1795, such as the Internet. The broadband modem 1798, which may be internal or external, is connected to the system bus 1714 via the serial port interface 1771. In a network environment, program modules related to the computer system 1700, or portions thereof, may be stored in the remote memory storage device 1790. Note that Fig.17 The network connections shown in are schematic and other means of establishing a communication link between the computers may be used depending on the specific requirements of the application of the current packet drop detection in the local networking stack through packet correlation.

[0068] Various exemplary embodiments of detecting packet drops in a local networking stack by packet correlation are now presented in an illustrative manner, rather than as an exhaustive list of all embodiments. The examples include a method performed by a computer server to monitor packet propagation and drops through a virtualized network stack hosted on the computer server, comprising: enabling operation of a packet monitoring application, the packet monitoring application triggering each component in the virtualized network stack to call an application programming interface (API) associated with the packet monitoring application; receiving parameters for filtering packets at the packet monitoring application; receiving packets at the packet monitoring application from a component within the virtualized network stack, wherein the component is a networking component configured to process and route packets through the virtualized network stack; filtering packets received from the component using the parameters by the packet monitoring application, wherein packets satisfying the parameters are selected for display on a user interface associated with the computer server; assigning an association ID (identifier) ​​to each packet at the packet monitoring application to enable association and tracking of packets as they traverse the components of the virtualized network stack.

[0069] In another example, the virtualized network stack includes a combination of hardware network devices and software defined network (SDN) components. In another example, the SDN components include a virtual switch, a virtual machine network interface controller (vNIC), a network driver interface specification (NDIS), and filters or extensions associated with the NDIS. In another example, the method also includes displaying a report of the filtered packets on a user interface. In another example, the method also includes encapsulating the packet with an external header and applying the filtering parameters to the internal header and the external header of the encapsulated packet. In another example, the method also includes using the assigned association ID to determine the delay of the packet at each component of the virtualized network stack. In another example, the component reports to the packet monitoring application the packet drops that occur at any point within the corresponding component, and the component reports successful packet propagation based on the corresponding packet that meets the component boundary. In another example, the parameter includes filtering the packet by the Internet Protocol (IP) address source. In another example, the parameter includes filtering the packet by the Internet Protocol (IP) address destination. In another example, the parameter includes filtering the packet by the MAC (media access control) address of the source or destination. In another example, the parameter includes filtering the packet by the Ethernet protocol. In another example, the parameters include filtering packets at various components of a plurality of available components within a virtualized network stack. In another example, the parameters include filtering packets by port source or port destination. In another example, the parameters include filtering packets by TCP (Transmission Control Protocol) flags, including any one or more of TCP SYN (Synchronize), TCP RST (Reset), or TCP FIN (Finish).

[0070] Another example includes a computer server configured with a packet monitoring application to track packets across a virtualized network stack on the computer server, the computer server comprising one or more processors and one or more hardware-based non-volatile memory devices storing computer-readable instructions that, when executed by the one or more processors, cause the computer server to: communicate packets between the computer server and one or more client computing devices, wherein during communication, packets traverse components of the virtualized network stack, the components being virtual network components or physical network components associated with the virtualized network stack; report the traversal status of the packets to the packet monitoring application by the components of the virtualized network stack; assign a relevant identifier (ID) to each packet; track packets across components on the virtualized network stack using the associated ID; and configure the packet monitoring application to filter packets based on set parameters so that, in response to packets satisfying one or more parameters, corresponding packets are selected for display on a user interface.

[0071] In another example, the filtering parameters are set in response to user input. In another example, the computer server further comprises a plurality of computer servers having respective packet monitoring applications for tracking packets, wherein the packets are tracked across a virtualized network stack of each computer server through which the packets traverse. In another example, the executed instructions further cause the computer server to display an option to enable and disable the packet monitoring application on a user interface associated with the computer server, wherein the component reports the status of the packet in response to the user interface receiving input to enable the packet monitoring application, and wherein a default setting of the packet monitoring application is disabled.

[0072] Another example includes one or more hardware-based non-transitory computer-readable storage devices storing instructions that, when executed by one or more processors disposed in a computer server, cause the computer server to: present a user interface to a user to enable a packet monitoring application, the packet monitoring application causing a component in a network stack to report packets; receive a selection of one or more parameters at the user interface that detail which packets are selected by the packet monitoring application; call an application programming interface (API) associated with the packet monitoring application by a component within the network stack, wherein the component uses the API to report packet drops and packet propagations to the packet monitoring application; receive packets from a component of the networking stack; use the parameters to filter the received packets, wherein packets that meet the parameters are selected by the packet monitoring application for display on a user interface associated with the computer server; assign unique correlation identifiers (IDs) to the filtered packets, wherein the correlation IDs assigned to the packets follow the corresponding packets through the network stack; calculate, by the packet monitoring application, delays for the packets at the component using information about the propagation of the packets from the component and the correlation IDs assigned to the individual packets; and display a report that includes the calculated delays and packet drops for the filtered packets.

[0073] In another example, the executed instructions further cause the computer server to register a component within the networking stack with the packet monitoring application, wherein only the registered component can call an API associated with the packet monitoring application.

[0074] Although the subject matter has been described in language specific to structural features and / or methodological acts, it is to be understood that the subject matter defined in the appended claims is not necessarily limited to the specific features or acts described above. Rather, the specific features and acts described above are disclosed as example forms of implementing the claims.

Claims

1. A method performed by a computer server to monitor packet propagation and discarding through a virtualized network stack hosted on the computer server, comprising: enabling operation of a packet monitoring application, the packet monitoring application triggering each component in a virtualized network stack to call an application programming interface API associated with the packet monitoring application, wherein the virtualized network stack includes components including a virtual switch, a physical network interface controller NIC, one or more virtual NICs, at least one network driver interface to one or more of the NICs, and an application socket API; receiving, at the packet monitoring application, parameters by which packets are filtered at each of the components in the virtualized network stack, wherein the computer server exposes a user interface by which the filtering parameters are selectable by a user; receiving, at the packet monitoring application, a packet from the component within the virtualized network stack, wherein the component is configured to process and route the packet traversing the virtualized network stack from the physical NIC to the application socket API, wherein the packet includes customer data traffic; filtering, by the packet monitoring application, packets received from the components using the filtering parameters, wherein the filtering is performed at each component that calls the API associated with the packet monitoring application during the packets' traversal of the virtualized network stack, and wherein packets satisfying the filtering parameters are selected for display on a user interface associated with the computer server; Encapsulate the packet with an outer header; applying the filter parameters to an inner header and an outer header of the encapsulated packet; as well as A correlation identifier ID is assigned to each packet at the packet monitoring application at the ingress of the physical NIC to enable correlation and tracking of packets as the packets traverse the components of the virtualized network stack from the physical NIC to the application socket API.

2. The method of claim 1, wherein the virtualized network stack comprises a combination of hardware networking devices and software-defined networking (SDN) components. 3 . The method according to claim 2 , wherein the SDN components include a virtual switch, a virtual machine network interface controller (vNIC), a network driver interface specification (NDIS), and filters or extensions associated with the NDIS. The method of claim 1 , further comprising displaying a report of the filtered groups on the user interface. 5 . The method of claim 1 , further comprising determining a latency of a packet at each component of the virtualized network stack using the assigned association ID.

6. The method according to claim 1, wherein: The components report packet drops occurring at any point within the respective components to the packet monitoring application, and The component reports successful packet propagation based on corresponding packets satisfying the component's boundaries.

7. The method of claim 1, wherein the parameters include filtering packets by Internet Protocol (IP) address source.

8. The method of claim 1, wherein the parameters include filtering packets by Internet Protocol (IP) address destination.

9. The method of claim 1, wherein the parameter comprises filtering packets by a Media Access Control (MAC) address of a source or destination.

10. The method of claim 1, wherein the parameter comprises filtering packets via an Ethernet protocol.

11. The method of claim 1, wherein the parameters include filtering packets at individual components of a plurality of available components within the virtualized network stack.

12. The method of claim 1, wherein the parameters include filtering packets by port source or port destination.

13. The method of claim 1, wherein the parameter comprises filtering packets by a transmission control protocol TCP flag, including any one or more of: TCP synchronization SYN, TCP reset RST, or TCP completion FIN.

14. A computer server configured with a packet monitoring application to track packets across a virtualized network stack on the computer server, comprising: one or more processors; a user interface configured to enable a user to select filtering parameters for the packets traversing the virtualized network stack; as well as one or more hardware-based non-transitory memory devices storing computer-readable instructions that, when executed by the one or more processors, cause the computer server to: communicating packets between the computer server and one or more client computing devices, wherein during the communication, the packets traverse components of the virtualized network stack, wherein the components include a virtual switch, a physical network interface controller NIC, one or more virtual NICs, at least one network driver interface to one or more of the NICs, and an application socket application programming interface (API), wherein the packets include client data traffic; reporting, by said component of said virtualized network stack, a traversal status of a packet to a packet monitoring application; Assigning an association identifier ID to each packet at the ingress of the physical NIC; configuring the packet monitoring application to filter packets according to selected filtering parameters, wherein the filtering is performed at each component that calls the API associated with the packet monitoring application during the packet's traversal of the virtualized network stack, and wherein in response to the packet satisfying one or more of the selected filtering parameters, the corresponding packet is selected for display on the user interface; Encapsulate the packet with an outer header; applying the filter parameters to an inner header and an outer header of the encapsulated packet; and The correlation ID is utilized to track filtered packets across components of the virtualized network stack from the physical NIC to the application socket API.

15. The computer server of claim 14, wherein the filtering parameters are set in response to user input from the user to the user interface.

16. The computer server of claim 14, further comprising a plurality of computer servers having respective packet monitoring applications for tracking packets, wherein packets are tracked across a virtualized network stack of each computer server over which the packets traverse.

17. The computer server of claim 14, wherein the instructions executed further cause the computer server to: displaying, on the user interface associated with the computer server, an option to enable and disable the packet monitoring application, in, The component reports a status of a packet in response to the user interface receiving input enabling the packet monitoring application, and wherein a default setting of the packet monitoring application is disabled.

18. One or more hardware-based non-transitory computer-readable memory devices storing instructions that, when executed by one or more processors disposed in a computer server, cause the computer server to: exposing a user interface to a user to enable a packet monitoring application that causes components within a virtualized network stack to report status of packets traversing the virtualized network stack, wherein the virtualized network stack includes components including a virtual switch, a physical network interface controller NIC, one or more virtual NICs, at least one network driver interface to one or more of the NICs, and an application socket application programming interface API; configuring the user interface to make one or more filtering parameters selectable by a user; calling, by the component within the virtualized network stack, a packet monitoring API, wherein the component uses the packet monitoring API to report packet drops and packet propagations to the packet monitoring application; receiving a packet from the component of the virtualized network stack, wherein the received packet includes customer data traffic; filtering the received packets using filtering parameters selected at each of the components in the virtualized network stack, wherein the filtering is performed at each component that calls the packet monitoring API during the packet's traversal of the virtualized network stack, and wherein packets that meet the filtering parameters are displayed on the user interface by the packet monitoring application; Encapsulate the packet with an outer header; applying the filter parameters to an inner header and an outer header of the encapsulated packet; assigning a unique correlation identifier ID to the filtered packets at the ingress of the physical NIC, wherein the correlation ID assigned to the packets follows the corresponding packets through the virtualized network stack from the physical NIC to the application socket API; calculating, by the packet monitoring application, a delay of the packet at the component using information about packet propagation from the component and an association ID assigned to the corresponding packet; as well as A report is displayed on the user interface, the report including the calculated delays and the packet drops for the filtered packets.

19. One or more hardware-based non-transitory computer-readable memory devices according to claim 18, wherein the executed instructions further cause the computer server to register components within the virtualized network stack with the packet monitoring application, wherein only registered components are able to call the API associated with the packet monitoring application.

Citation Information

Patent Citations

  • Automated correlation and analysis of callstack and context data

    CN105144117A

  • Virtual Network Flow Monitoring

    US20140281030A1