Computing power resource network virtualization method and device of intelligent computing center

Abstract

The invention provides a computing power resource network virtualization method and device of an intelligent computing center, and relates to the technical field of computing power infrastructure, and the method comprises the steps: carrying out abstraction processing on computing power resources, and forming a virtual resource pool; in the virtual resource pool, a plurality of virtual networks are created, and the plurality of virtual networks are isolated from one another; configuring corresponding network parameters for each virtual network according to the service type of the computing power service borne by each virtual network; and adjusting network parameters of each virtual network according to the running state data of each virtual network. Therefore, computing power resources are abstracted to form a virtual resource pool, mutually isolated virtual networks are created, network parameters are configured according to service types, and the parameters are adjusted according to the operation state data, so that resource integration can be realized, service isolation safety can be guaranteed, service requirements can be accurately adapted, the networks can be dynamically optimized, and the computing power resource utilization efficiency and the network performance can be improved.

Description

Technical Field

[0001] The present invention relates to the technical field of intelligent computing centers, smart computing centers and computing power infrastructure, and specifically to a method and device for network virtualization of computing power resources in an intelligent computing center. Background Art

[0002] With the rapid development of artificial intelligence technology, "intelligent computing centers" and "intelligent computing centers" have emerged.

[0003] An "Intelligent Computing Center" is a facility that uses large-scale heterogeneous computing resources, including general-purpose and intelligent computing power, to provide the computing power, data, and algorithms required for AI applications (such as AI deep learning model development, model training, and model inference). The Intelligent Computing Center encompasses facilities, hardware, and software, and provides a full stack of capabilities, from bottom-level computing power to top-level application enablement.

[0004] “Intelligent Computing Center” includes but is not limited to “Smart Computing Center”.

[0005] "Intelligent Computing Center" refers to an artificial intelligence computing center. It is a type of computing power infrastructure that is based on artificial intelligence theory, adopts artificial intelligence computing architecture, and provides computing power services, data services, and algorithm services required for artificial intelligence applications.

[0006] "Computing power" is the core of "intelligent computing center" and "intelligent computing center". It is the ability of computer equipment or computing / data center to process information. It is the ability of computer hardware and software to work together to perform certain computing needs. It is the computing power to achieve target result output by processing information data. It is a new type of productivity that integrates information computing power, network carrying capacity, and data storage capacity. It mainly provides services to society through computing power infrastructure.

[0007] As the core hub for processing massive amounts of data and performing complex computing tasks, intelligent computing centers are tasked with providing powerful computing power to numerous enterprises, research institutions, and various intelligent applications. Within intelligent computing centers, numerous physical devices operate independently, requiring device-specific operations and maintenance. This creates a heavy workload and high costs, making it difficult to uniformly allocate and optimize computing resources. Furthermore, the failure of a single physical device can disrupt related services, and the lack of effective fault tolerance mechanisms and rapid recovery methods can impact business continuity. Therefore, since the emergence of intelligent computing centers, achieving network virtualization of computing resources has become a pressing technical challenge. Summary of the Invention

[0008] The present invention provides a method and device for network virtualization of computing power resources of an intelligent computing center to solve the problem of how to realize network virtualization of computing power resources.

[0009] To solve the above problems, the present invention is achieved as follows:

[0010] In a first aspect, the present invention provides a method for network virtualization of computing resources in an intelligent computing center, comprising:

[0011] Step S1: abstract computing resources to form a virtual resource pool;

[0012] Step S2: creating multiple virtual networks in the virtual resource pool, wherein the multiple virtual networks are isolated from each other;

[0013] Step S3: configuring corresponding network parameters for each virtual network according to the service type of the computing power service carried by each virtual network;

[0014] Step S4: Adjust the network parameters of each virtual network according to the operation status data of each virtual network.

[0015] In one embodiment, the multiple virtual networks include a virtual local area network (VLAN), and step S2 includes:

[0016] Step S21: Determine the identification information of the VLAN;

[0017] Step S22: creating identification information of the VLAN on the switch, and configuring a port group of the switch corresponding to the identification information of the VLAN in the virtual resource pool;

[0018] Step S23: Create an access control list (ACL) based on the port group of the switch corresponding to the identification information of the VLAN.

[0019] In one embodiment, the multiple virtual networks include a virtual private network (VPN), and step S2 includes:

[0020] Step S24: Determine the VPN protocol type, encryption algorithm, and tunnel parameters.

[0021] Step S25: deploying a server instance and a client instance in the virtual resource pool based on the protocol type of the VPN;

[0022] Step S26: establishing a tunnel connection between the server instance and the client instance based on the tunnel parameters;

[0023] Step S27: Determine the encryption strategy of the tunnel based on the encryption algorithm.

[0024] In one embodiment, the network parameters include network topology parameters, and the network topology parameters include a connection mode between network nodes of the virtual network and a network hierarchical structure;

[0025] The step S3 comprises:

[0026] Step S31: Determine the service characteristics and service scale of the computing service carried by each virtual network according to the service type of the computing service carried by the virtual network;

[0027] Step S32: determining the number of interactions between the network nodes according to the service characteristics;

[0028] Step S33: determining a connection mode between the network nodes based on the number of interactions between the network nodes, wherein if the number of interactions between the network nodes is greater than a preset number of interactions, the connection mode between the network nodes is a direct connection; and if the number of interactions between the network nodes is less than or equal to the preset number of interactions, the connection mode between the network nodes is an indirect connection;

[0029] Step S34: Determine the network hierarchy structure based on the business scale, wherein, when the business scale is greater than the preset business scale, the network hierarchy structure is a multi-level network structure; when the business scale is less than or equal to the preset business scale, the network hierarchy structure is a flat network structure.

[0030] In one embodiment, the service characteristics include data traffic size, delay sensitivity, data integrity requirements, and service priority; the network parameters include network performance parameters, and the network performance parameters include network bandwidth requirements, network delay targets, network packet loss rates, and QoS policies; and step S3 further includes:

[0031] Step S35: Determine the network bandwidth requirement based on the data traffic size;

[0032] Step S36: Determine the network delay target based on the delay sensitivity;

[0033] Step S37: determining the network packet loss rate based on the data integrity requirement;

[0034] Step S38: Determine the QoS policy based on the service priority.

[0035] In one embodiment, step S4 includes:

[0036] Step S41: Calculate resource demand data of each virtual network using a pre-trained machine learning model based on the operating status data of each virtual network;

[0037] Step S42: Adjust the network parameters of each virtual network according to the resource demand data of each virtual network.

[0038] In a second aspect, the present invention further provides a computing resource network virtualization device for an intelligent computing center, comprising:

[0039] The first processing module is used to abstract computing resources to form a virtual resource pool;

[0040] A first creation module is used to create multiple virtual networks in the virtual resource pool, where the multiple virtual networks are isolated from each other;

[0041] A first configuration module is used to configure corresponding network parameters for each virtual network according to the service type of the computing power service carried by each virtual network;

[0042] The first adjustment module is configured to adjust the network parameters of each virtual network according to the operation status of each virtual network.

[0043] In a third aspect, the present invention also provides an electronic device comprising a processor, a memory, and a computer program stored in the memory and executable on the processor. When the computer program is executed by the processor, the steps in the method for network virtualization of computing resources of an intelligent computing center as described in the first aspect above are implemented.

[0044] In a fourth aspect, the present invention also provides a computer-readable storage medium, on which a computer program is stored. When the computer program is executed by a processor, the steps in the method for network virtualization of computing resources of an intelligent computing center as described in the first aspect above are implemented.

[0045] In a fifth aspect, the present invention further provides a computer program product comprising computer instructions, which, when executed by a processor, implement the steps in the method for network virtualization of computing resources of an intelligent computing center as described in the first aspect above.

[0046] In the present invention, computing power resources are abstracted to form a virtual resource pool; in the virtual resource pool, multiple virtual networks are created, and the multiple virtual networks are isolated from each other; according to the business type of the computing power business carried by each virtual network, corresponding network parameters are configured for each virtual network; according to the operating status data of each virtual network, the network parameters of each virtual network are adjusted. In this way, by abstracting computing power resources to form a virtual resource pool, creating mutually isolated virtual networks, configuring network parameters according to business types, and adjusting parameters according to operating status data, it is possible to achieve resource integration, ensure business isolation security, accurately adapt to business needs, dynamically optimize the network, and improve computing power resource utilization efficiency and network performance. BRIEF DESCRIPTION OF THE DRAWINGS

[0047] In order to more clearly illustrate the technical solution of the present invention, the following is a brief introduction to the drawings required for the description of the present invention. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying any creative labor.

[0048] Figure 1 This is a flow chart of a method for network virtualization of computing resources in an intelligent computing center provided by the present invention;

[0049] Figure 2 This is one of the schematic diagrams of the QoS strategy of the virtual network provided by the present invention;

[0050] Figure 3 This is the second schematic diagram of the QoS policy of the virtual network provided by the present invention;

[0051] Figure 4 This is the third schematic diagram of the QoS policy of the virtual network provided by the present invention;

[0052] Figure 5 This is a structural diagram of a computing power resource network virtualization device for an intelligent computing center provided by the present invention;

[0053] Figure 6 This is a structural diagram of an electronic device provided by the present invention. DETAILED DESCRIPTION

[0054] The following will clearly and completely describe the technical solutions of the present invention in conjunction with the accompanying drawings. Obviously, the embodiments described are only some embodiments of the present invention, not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without making any creative efforts shall fall within the scope of protection of the present invention.

[0055] The "computing power" mentioned in the present invention refers to: the ability of computer equipment or computing / data centers to process information, the ability of computer hardware and software to work together to execute certain computing requirements, and the computing power to achieve target result output by processing information data. It is a new type of productivity that integrates information computing power, network carrying capacity, and data storage capacity, and mainly provides services to society through computing power infrastructure.

[0056] The "computing power" (Computational Power, CP) mentioned in the present invention refers to: the ability of a data center server to process data and output results. It is a comprehensive indicator to measure the computing power of a data center, including general computing power, super computing power and intelligent computing power. The commonly used unit of measurement is the number of floating-point operations performed per second (FLOPS, 1EFLOPS=10^18FLOPS). The larger the value, the stronger the comprehensive computing power. According to calculations, 1EFLOPS is approximately the computing power output of 5 Tianhe-2A or 500,000 mainstream server CPUs or 2 million mainstream notebooks. The calculation formula is: CP=CP 通用 +CP 智能 +CP 超级 .

[0057] The "carrying capacity" (Network Power, NP) mentioned in the present invention refers to: it is the performance of the data transmission capability of the computing power facility, which includes comprehensive capabilities such as network architecture, network bandwidth, transmission latency, intelligent management and scheduling, etc. It involves network transmission within and between data centers, and is a comprehensive indicator for measuring network transmission scheduling capabilities.

[0058] The "Storage Power" (SP) described in this invention refers to the comprehensive capabilities of a data center in terms of data storage capacity, performance, security and reliability, and environmental friendliness. It is a comprehensive indicator for measuring a data center's data storage capacity, encompassing both external storage devices such as storage arrays and server-internal storage. Storage capacity is commonly measured in exabytes (EB, 1EB = 2^60 bytes), while performance is commonly measured in IOPS / TB (Input / Output Operations Per Second / TB). Disaster recovery ratio is a key indicator of security and reliability.

[0059] The "computing power infrastructure" mentioned in the present invention refers to a new type of information infrastructure that integrates information computing power, network carrying capacity, and data storage capacity, and can realize the centralized calculation, storage, transmission and application of information.

[0060] The "new information infrastructure" mentioned in the present invention refers to: mainly including network infrastructure such as 5G networks, fiber-optic broadband networks, backbone networks, international communication networks, satellite Internet, computing power infrastructure such as data centers, general computing power centers, intelligent computing centers, supercomputing centers, and new technology facilities such as artificial intelligence, blockchain, and quantum computing.

[0061] The "computing power" mentioned in the present invention includes: general computing power, intelligent computing power and super computing power.

[0062] The "general computing power" mentioned in the present invention refers to the computing power provided by servers based on central processing unit (CPU) chips, which is used to support basic general computing such as cloud computing and edge computing.

[0063] The "intelligent computing power" mentioned in the present invention refers to: a computing platform based on large-scale deployment of special chips such as graphics processing units (GPUs), field programmable gate arrays (FPGAs), and application-specific integrated circuits (ASICs) for various innovative artificial intelligence applications, such as natural language processing and machine vision.

[0064] The "supercomputing power" mentioned in the present invention refers to the computing power provided by high-performance computing clusters such as supercomputers. It utilizes the centralized computing resources of multiple computer systems working in parallel and uses a dedicated operating system to handle extremely complex or data-intensive problems. It is mainly used for calculations in cutting-edge scientific fields, such as planetary simulation, drug molecule design, genetic analysis, etc.

[0065] The "intelligent computing center" described in this article refers to a facility that provides the computing power, data, and algorithms required for artificial intelligence applications (such as AI deep learning model development, model training, and model inference) by utilizing large-scale heterogeneous computing resources, including general-purpose computing power (CPU) and intelligent computing power (GPU, FPGA, ASIC, etc.). The intelligent computing center encompasses facilities, hardware, and software, and can provide a full stack of capabilities, from bottom-level computing power to top-level application enablement.

[0066] The "intelligent computing center" mentioned in the present invention includes but is not limited to the "intelligent computing center".

[0067] The "intelligent computing center" mentioned in the present invention is an artificial intelligence computing center, which is a type of computing power infrastructure based on artificial intelligence theory, adopts artificial intelligence computing architecture, and provides computing power services, data services and algorithm services required for artificial intelligence applications.

[0068] The "computing power center" mentioned in the present invention refers to: a facility that is mainly composed of infrastructure such as wind, fire, water, electricity, and IT hardware and software equipment, and has computing power, transportation capacity, and storage capacity, including general data centers, intelligent computing centers, supercomputing centers, etc.

[0069] The "supercomputing center" mentioned in the present invention refers to: a supercomputing data center, which is a data center based on a supercomputer or a large-scale computing cluster, which can provide large-scale computing, storage and network services and other functions, and is widely used in application scenarios such as aerospace, national defense, oil exploration, climate modeling and genome sequencing.

[0070] The "computing resources" mentioned in the present invention refer to: technologies and facilities with information computing, transmission, storage and application capabilities required for the development of a digital society, including but not limited to computing resources such as CPUs and GPUs, network resources such as switches and routers, storage resources such as storage arrays and distributed storage, security resources such as firewalls and intrusion detection systems, and supporting and guarantee resources such as wind, fire, water and electricity.

[0071] Network virtualization, as used in this article, refers to the use of software to abstract physical network resources (such as bandwidth, switches, and routers) into multiple independent logical networks. This allows the creation of multiple virtual networks on a single physical network, each with its own address space, topology, and security policies, ensuring isolation and non-interference.

[0072] In the prior art, in intelligent computing centers, numerous physical devices operate independently, and operation and maintenance need to be carried out for different devices. The management workload is large, the cost is high, and it is difficult to achieve unified allocation and optimization of resources. In addition, the failure of a single physical device may cause the interruption of related services. The lack of effective fault tolerance mechanisms and rapid recovery methods will affect the continuity of services. Therefore, since the emergence of intelligent computing centers, how to achieve network virtualization of computing power resources has become a technical problem that needs to be solved urgently. In order to achieve network virtualization of computing power resources of intelligent computing centers, in the present invention, computing power resources are abstracted to form a virtual resource pool; in the virtual resource pool, multiple virtual networks are created, and the multiple virtual networks are isolated from each other; according to the business type of the computing power business carried by each virtual network, corresponding network parameters are configured for each virtual network; according to the operating status data of each virtual network, the network parameters of each virtual network are adjusted. In this way, computing resources are abstracted to form a virtual resource pool, isolated virtual networks are created, network parameters are configured according to business types, and parameters are adjusted according to operating status data. This can achieve resource integration, ensure business isolation security, accurately adapt to business needs, dynamically optimize the network, and improve computing resource utilization efficiency and network performance.

[0073] For details, see Figure 1 , Figure 1 This is a flow chart of a method for network virtualization of computing resources in an intelligent computing center provided by the present invention. Figure 1 As shown, the following steps are included:

[0074] Step S1: abstract computing resources to form a virtual resource pool;

[0075] In this step, various types and specifications of physical computing resources, such as server CPUs, memory, and storage, are abstracted and integrated through software technology. This allows them to be no longer constrained by specific physical hardware, but to form a unified, flexibly deployable virtual resource pool. This allows users or applications to see not individual physical devices, but a collection of virtual resources with a unified interface and management method.

[0076] Step S2: creating multiple virtual networks in the virtual resource pool, wherein the multiple virtual networks are isolated from each other;

[0077] In this step, virtualization technology is used to create multiple independent virtual networks within the existing virtual resource pool. These virtual networks are logically isolated from each other, acting as independent network environments. Traffic between different virtual networks cannot directly communicate with each other, and each has its own independent network topology, IP address space, and other features.

[0078] Step S3: configuring corresponding network parameters for each virtual network according to the service type of the computing power service carried by each virtual network;

[0079] In this step, specific computing services carried by each virtual network are configured, such as big data processing services and real-time video streaming services. Based on the different network requirements of these computing services, corresponding network parameters such as bandwidth, latency, and packet loss rate are configured for the virtual network. For example, for a virtual network carrying real-time video streaming, higher bandwidth and lower latency requirements may be configured to ensure smooth video playback.

[0080] Step S4: Adjust the network parameters of each virtual network according to the operation status data of each virtual network.

[0081] In this step, the operational status of each virtual network is monitored, such as real-time bandwidth usage, latency changes, and packet loss. The network parameters of each virtual network are dynamically adjusted based on this actual operational data. For example, if bandwidth usage is excessively high for a particular virtual network, its available bandwidth is appropriately increased. Or, if latency is abnormal, network paths are optimized to reduce latency. It should be noted that each virtual network's network parameter adjustments are based on its own network's operational status data.

[0082] In the above embodiment, computing power resources are abstracted to form a virtual resource pool; in the virtual resource pool, multiple virtual networks are created, and the multiple virtual networks are isolated from each other; according to the business type of the computing power business carried by each virtual network, corresponding network parameters are configured for each virtual network; according to the operating status data of each virtual network, the network parameters of each virtual network are adjusted. In this way, by abstracting computing power resources to form a virtual resource pool, creating mutually isolated virtual networks, configuring network parameters according to business types, and adjusting parameters according to operating status data, resource integration can be achieved, business isolation security can be guaranteed, business needs can be accurately adapted, the network can be dynamically optimized, and computing power resource utilization efficiency and network performance can be improved.

[0083] In one embodiment, the multiple virtual networks include a virtual local area network (VLAN), and step S2 includes:

[0084] Step S21: Determine the identification information of the VLAN;

[0085] Step S22: creating identification information of the VLAN on the switch, and configuring a port group of the switch corresponding to the identification information of the VLAN in the virtual resource pool;

[0086] Step S23: Create an access control list (ACL) based on the port group of the switch corresponding to the identification information of the VLAN.

[0087] In the above embodiment, when creating a virtual local area network (VLAN), the first step is to clearly define the VLAN division requirements, group network users by department, business, or function, and determine the identification information for each VLAN to facilitate subsequent identification and management within network devices. A VLAN identification is typically a 12-bit number ranging from 1 to 4094. Administrators can assign different identification information to different VLANs based on network planning and business requirements.

[0088] Next, log in to the switch using the command line or graphical user interface. In global configuration mode, create a VLAN using a command such as "vlan [VLAN ID]" and give it a name. Then, enter switch port configuration mode and assign the switch ports to the appropriate VLANs. Create a port group corresponding to the VLAN in the virtual resource pool so that virtual machines can connect to the correct VLAN network.

[0089] Finally, access control rules are created for network traffic in a specific VLAN to restrict or allow network access between different VLANs and within a VLAN, enhancing network security and manageability. An Access Control List (ACL) is a set of rules that determines whether to allow or deny a packet based on information such as the packet's source IP address, destination IP address, port number, and protocol type. Based on the switch's port group (i.e., the set of ports belonging to a specific VLAN), inbound and outbound ACL rules can be created. For example, rules can be created to prohibit hosts in VLAN 100 from accessing certain ports of a specific server in VLAN 200, or to allow communication of a specific protocol (such as HTTP) between hosts within VLAN 100.

[0090] Furthermore, when different VLANs need to communicate, inter-VLAN routing is required. This can be achieved through single-arm routing using a router's sub-interfaces, or through the Switched Virtual Interface (SVI) of a Layer 3 switch. For example, on a Layer 3 switch, create an SVI interface for each VLAN and configure the corresponding IP address. Hosts in different VLANs can then communicate through these SVI interfaces.

[0091] To transmit traffic from multiple VLANs between switches or between a switch and a router, use a trunk link. A trunk link tags traffic from different VLANs, allowing a single physical link to carry traffic from multiple VLANs. When configuring a trunk link, ensure that the trunk mode and the list of allowed VLANs on both devices are consistent.

[0092] In addition to port-based VLAN division, VLAN division can also be based on Media Access Control (MAC) addresses, IP subnets, protocol types, etc. For example, VLAN division based on MAC addresses ensures that no matter which port a device is connected to a switch, it belongs to a specific VLAN, which is suitable for scenarios with high device mobility requirements.

[0093] In the above embodiment, establishing VLANs improves network security. By assigning different departments or users to different VLANs, the broadcast domain is limited, reducing the risk of information leakage. Furthermore, VLANs enhance network performance, reduce the broadcast range, and minimize network congestion. Furthermore, VLANs facilitate network management, making the movement, addition, and modification of users or devices more flexible, without requiring major adjustments to the network architecture.

[0094] In one embodiment, the multiple virtual networks include a virtual private network (VPN), and step S2 includes:

[0095] Step S24: Determine the VPN protocol type, encryption algorithm, and tunnel parameters.

[0096] Step S25: deploying a server instance and a client instance in the virtual resource pool based on the protocol type of the VPN;

[0097] Step S26: establishing a tunnel connection between the server instance and the client instance based on the tunnel parameters;

[0098] Step S27: Determine the encryption strategy of the tunnel based on the encryption algorithm.

[0099] In the above embodiments, the protocol type of the Virtual Private Network (VPN) determines the communication rules and functional characteristics of the VPN. For example, OpenVPN has strong compatibility, and Internet Protocol Security (IPSec) is suitable for enterprise networks. Selecting the appropriate protocol is the basis for network operation. Encryption algorithms are used to ensure data transmission security. For example, the Advanced Encryption Standard (AES) algorithm can effectively prevent data theft and tampering. Tunnel parameters include the tunnel interface address and the Maximum Transmission Unit (MTU). These parameters affect the performance and stability of the VPN network.

[0100] Based on the selected protocol type, computing and storage resources are allocated within the virtual resource pool to deploy the corresponding server and client. Different protocols have different server and client configuration requirements. For example, an OpenVPN server requires certificate services and listening ports, while clients require the corresponding adapter software. Deployment instances provide the physical support for VPN connections.

[0101] Using the configured tunnel parameters, a data transmission channel is established between the server and the client. For example, by configuring the tunnel interface address and routing rules, client data can accurately enter the tunnel and be forwarded by the server, achieving secure data transmission. The tunnel connection is the core link for remote access in VPN networks.

[0102] Then, based on the selected encryption algorithm, a specific encryption strategy is developed, including key generation, management, and update methods, as well as the data encryption and decryption process. This ensures that data remains encrypted during tunnel transmission. Even if the data is intercepted, attackers cannot obtain the actual content, thus ensuring the security and integrity of data transmission.

[0103] In the above examples, establishing a VPN can enhance network security by encrypting user data traffic, protecting it from hackers or malware. Furthermore, a VPN can hide a user's real IP address, making their online activities more anonymous and reducing the risk of being tracked.

[0104] In one embodiment, the network parameters include network topology parameters, and the network topology parameters include a connection mode between network nodes of the virtual network and a network hierarchical structure;

[0105] The step S3 comprises:

[0106] Step S31: Determine the service characteristics and service scale of the computing service carried by each virtual network according to the service type of the computing service carried by the virtual network;

[0107] Step S32: determining the number of interactions between the network nodes according to the service characteristics;

[0108] Step S33: determining a connection mode between the network nodes based on the number of interactions between the network nodes, wherein if the number of interactions between the network nodes is greater than a preset number of interactions, the connection mode between the network nodes is a direct connection; and if the number of interactions between the network nodes is less than or equal to the preset number of interactions, the connection mode between the network nodes is an indirect connection;

[0109] Step S34: Determine the network hierarchy structure based on the business scale, wherein, when the business scale is greater than the preset business scale, the network hierarchy structure is a multi-level network structure; when the business scale is less than or equal to the preset business scale, the network hierarchy structure is a flat network structure.

[0110] In the above embodiments, different types of computing services, such as data processing, AI training, and graphics rendering, have different service characteristics and scales. For example, AI training is characterized by being computationally intensive, with a large data volume and high computing resource requirements.

[0111] Service characteristics influence the interactions between network nodes. For example, data-intensive services may require frequent transmission of large amounts of data between different nodes, resulting in a high number of interactions; whereas compute-intensive services may have relatively few interactions. Service characteristics can be used to analyze the number of interactions between network nodes and further inform network connectivity planning.

[0112] The connection method is determined based on the number of interactions between network nodes. If the number of interactions is greater than a preset number, it indicates that communication between these nodes is frequent. Using a direct connection can reduce forwarding delays at intermediate nodes and improve communication efficiency. If the number of interactions is less than or equal to the preset number, using an indirect connection can save network resources and avoid unnecessary direct connections that waste resources.

[0113] The scale of the business determines the network's hierarchical structure. When the business scale exceeds the preset business scale, the network needs to process a large amount of data and a large number of nodes. A multi-level network structure can better manage resources and traffic, implement different functions through layering, and improve network scalability and stability. When the business scale is smaller, a flat network structure is simple and direct, which can reduce network complexity and improve network flexibility and responsiveness.

[0114] In the above embodiments, optimizing the network connection method and hierarchical structure based on service characteristics and scale can better adapt the network to the needs of different computing services, reduce data transmission delays, and improve the overall performance and efficiency of the network. Properly selecting the connection method and network hierarchy can avoid over-allocation or under-allocation of resources, achieve efficient utilization of network resources, and reduce costs.

[0115] In one embodiment, the service characteristics include data traffic size, delay sensitivity, data integrity requirements, and service priority; the network parameters include network performance parameters, and the network performance parameters include network bandwidth requirements, network delay targets, network packet loss rates, and QoS policies; and step S3 further includes:

[0116] Step S35: Determine the network bandwidth requirement based on the data traffic size;

[0117] Step S36: Determine the network delay target based on the delay sensitivity;

[0118] Step S37: determining the network packet loss rate based on the data integrity requirement;

[0119] Step S38: Determine the QoS policy based on the service priority.

[0120] In the above embodiments, the data traffic volumes of different computing services vary significantly. For example, data-intensive services such as video streaming and big data analysis generate large amounts of data transmission, requiring a large network bandwidth to ensure smooth data transmission and avoid data congestion and delays. On the other hand, simple text processing or small computing tasks have relatively small data traffic volumes and lower network bandwidth requirements. By determining network bandwidth requirements based on data traffic volume, appropriate network transmission capacity can be provided to each virtual network, ensuring the normal operation of the service.

[0121] Latency sensitivity refers to a service's tolerance for data transmission delays. For example, services like real-time video conferencing and autonomous driving are extremely latency-sensitive. Even small delays can lead to serious consequences, such as video freezes and vehicle control errors. Therefore, network latency targets must be set very low to ensure timely data transmission and processing. Non-real-time services, such as batch data processing and file downloads, have relatively low latency requirements and can tolerate a certain degree of delay. Determining network latency targets based on service latency sensitivity can meet the time requirements of different services.

[0122] Data integrity requirements refer to the degree to which a business requires data to be free of loss or errors during transmission. For example, businesses like financial transactions and medical data transmission have extremely high data integrity requirements. Any data loss can have serious consequences, so network packet loss rates must be kept to a very low level. On the other hand, for businesses with relatively low data integrity requirements, such as standard web browsing and online music streaming, basic business functions can be maintained despite a certain degree of packet loss. Determining the network packet loss rate based on data integrity requirements can ensure the accuracy and reliability of business data.

[0123] Business priorities reflect the importance of different businesses within the entire system. For example, an enterprise's core business systems and mission-critical computing tasks typically have higher priorities, while auxiliary businesses or non-critical tasks have lower priorities. By determining business priorities, Quality of Service (QoS) policies can be formulated accordingly. Figure 2 and Figure 3 This is a diagram of creating a QoS policy in a virtual network. By creating a QoS policy, more network resources can be allocated to high-priority services, such as priority transmission and guaranteed bandwidth. This ensures that critical services are given priority and can operate normally even when network resources are tight, while low-priority services can use the remaining network resources without affecting high-priority services.

[0124] In the above steps, accurately configuring network parameters according to the specific characteristics of different computing power services can ensure the reasonable allocation of network resources, avoid resource waste or shortage, thereby better meeting the operating requirements of various services and improving service processing efficiency and quality.

[0125] In one embodiment, step S4 includes:

[0126] Step S41: Calculate resource demand data of each virtual network using a pre-trained machine learning model based on the operating status data of each virtual network;

[0127] Step S42: Adjust the network parameters of each virtual network according to the resource demand data of each virtual network.

[0128] In the above embodiment, each virtual network will generate various operating status data during operation, such as the actual usage of network bandwidth, real-time value of network delay, CPU and memory utilization, etc. These data can reflect the current operating status of the virtual network and the resource usage pattern. The pre-trained machine learning model is obtained by training with a large amount of historical data. These historical data contain the operating status data of the virtual network in different scenarios and the corresponding resource requirements. The model learns the potential relationship between operating status data and resource requirements. Using this machine learning model, the operating status data of the current virtual network is used as input, and the model can calculate the resource demand data of the virtual network in the future, such as calculating the amount of network bandwidth that may need to be increased, whether the number of CPU cores needs to be expanded, etc.

[0129] If you calculate that the network bandwidth demand will increase, you can increase the network bandwidth allocation of the virtual network accordingly; if you calculate that the CPU resources are tight, you may need to adjust the configuration of the virtual machine, increase the number of CPU cores, or adjust the distribution of tasks to avoid service performance degradation due to insufficient resources. Figure 4 , when the running status of the virtual network changes, you can Figure 3 Adjust the bandwidth based on the configured QoS policy to 2 Mb.

[0130] In the above embodiments, by calculating resource demand data, future resource requirements for the virtual network can be understood in advance, allowing for pre-emptive resource planning and adjustment. This avoids the need to wait until resource shortages are discovered and urgently addressed, reduces the risk of performance degradation or interruption due to resource bottlenecks, and enables more rational and optimized resource allocation.

[0131] See Figure 5 , Figure 5 This is a structural diagram of a computing power resource network virtualization device for an intelligent computing center provided by the present invention, such as Figure 5 As shown, the computing power resource network virtualization device 600 of the intelligent computing center includes:

[0132] The first processing module 501 is used to abstract computing resources to form a virtual resource pool;

[0133] A first creation module 502 is configured to create multiple virtual networks in the virtual resource pool, wherein the multiple virtual networks are isolated from each other;

[0134] A first configuration module 503 is configured to configure corresponding network parameters for each virtual network according to the service type of the computing power service carried by each virtual network;

[0135] The first adjustment module 504 is configured to adjust the network parameters of each virtual network according to the operation status of each virtual network.

[0136] In one embodiment, the multiple virtual networks include a virtual local area network (VLAN), and the first creation module includes:

[0137] A first determining unit, configured to determine identification information of a VLAN;

[0138] A first configuration unit, configured to create identification information of the VLAN on the switch, and configure a port group of the switch corresponding to the identification information of the VLAN in the virtual resource pool;

[0139] The first creating unit is configured to create an access control list (ACL) based on the port group of the switch corresponding to the identification information of the VLAN.

[0140] In one embodiment, the multiple virtual networks include a virtual private network (VPN), and the first creation module includes:

[0141] A second determining unit is used to determine the protocol type, encryption algorithm and tunnel parameters of the VPN;

[0142] A first deployment unit, configured to deploy a server instance and a client instance in the virtual resource pool based on a protocol type of the VPN;

[0143] a first establishing unit, configured to establish a tunnel connection between the server instance and the client instance based on the tunnel parameters;

[0144] The third determining unit is configured to determine an encryption policy for the tunnel based on the encryption algorithm.

[0145] In one embodiment, the network parameters include network topology parameters, and the network topology parameters include a connection mode between network nodes of the virtual network and a network hierarchical structure;

[0146] The first configuration module includes:

[0147] A fourth determining unit, configured to determine the service characteristics and service scale of the computing service carried by each virtual network according to the service type of the computing service carried by the virtual network;

[0148] a fifth determining unit, configured to determine a number of interactions between the network nodes according to the service characteristics;

[0149] a sixth determining unit, configured to determine a connection mode between the network nodes based on a number of interactions between the network nodes, wherein if the number of interactions between the network nodes is greater than a preset number of interactions, the connection mode between the network nodes is a direct connection; and if the number of interactions between the network nodes is less than or equal to the preset number of interactions, the connection mode between the network nodes is an indirect connection;

[0150] The seventh determination unit is used to determine the network hierarchical structure according to the business scale, wherein, when the business scale is greater than the preset business scale, the network hierarchical structure is a multi-level network structure, and when the business scale is less than or equal to the preset business scale, the network hierarchical structure is a flat network structure.

[0151] In one embodiment, the service characteristics include data traffic size, delay sensitivity, data integrity requirements, and service priority; the network parameters include network performance parameters, and the network performance parameters include network bandwidth requirements, network delay targets, network packet loss rates, and QoS policies; and the first configuration module further includes:

[0152] an eighth determining unit, configured to determine the network bandwidth requirement based on the data traffic size;

[0153] a ninth determining unit, configured to determine the network delay target based on the delay sensitivity;

[0154] a tenth determining unit, configured to determine the network packet loss rate based on the data integrity requirement;

[0155] An eleventh determining unit is configured to determine the QoS policy based on the service priority.

[0156] In one embodiment, the first adjustment module includes:

[0157] A first calculation unit is configured to calculate resource demand data of each virtual network using a pre-trained machine learning model according to the operation status data of each virtual network;

[0158] The first adjusting unit is configured to adjust the network parameters of each virtual network according to the resource demand data of each virtual network.

[0159] The computing power resource network virtualization device for the intelligent computing center provided by the present invention is capable of realizing the various processes of each embodiment of the computing power resource network virtualization method of the above-mentioned intelligent computing center. The technical features correspond one to one and can achieve the same technical effects. To avoid repetition, they will not be described here.

[0160] It should be noted that the computing power resource network virtualization device of the intelligent computing center in the present invention can be a device, or a component, integrated circuit, or chip in an electronic device.

[0161] The present invention also provides an electronic device, see Figure 6 , Figure 6 The electronic device includes a memory 601, a processor 602, and a program or instruction stored in the memory 601 and executed by the processor 602. Figure 1 Any steps in the corresponding embodiment of the method for network virtualization of computing resources of the intelligent computing center and the same beneficial effects are achieved will not be repeated here.

[0162] The processor 602 may be a CPU, an ASIC, an FPGA, or a GPU.

[0163] Those skilled in the art will understand that all or part of the steps of the embodiment of the method for implementing the network virtualization of computing resources of the above-mentioned intelligent computing center can be completed through hardware related to program instructions, and the program can be stored in a readable medium.

[0164] The present invention also provides a readable storage medium, which stores a computer program. When the computer program is executed by a processor, the above Figure 1 Any step in the corresponding embodiment of the method for virtualizing computing resources in a network of an intelligent computing center can achieve the same technical effect and will not be described here to avoid repetition. The storage medium is, for example, a read-only memory (ROM), a random access memory (RAM), a magnetic disk, or an optical disk.

[0165] The present invention also provides a computer program product, comprising computer instructions, which, when executed by a processor, implement the above Figure 1 The various processes of the implementation method of the computing power resource network virtualization method of the corresponding intelligent computing center can achieve the same technical effect. To avoid repetition, they will not be repeated here.

[0166] The terms "first", "second" and the like in the present invention are used to distinguish similar objects and are not necessarily used to describe a specific order or sequential order. In addition, the terms "comprise" and "have" and any of their variations are intended to cover non-exclusive inclusions, for example, a process, method, system, product or device comprising a series of steps or units is not necessarily limited to those steps or units clearly listed, but may include other steps or units that are not clearly listed or that are inherent to these processes, methods, products or devices. In addition, "and / or" is used in this application to represent at least one of the connected objects, for example A and / or B and / or C, which means comprising seven situations including single A, single B, single C, and both A and B exist, both B and C exist, both A and C exist, and both A, B and C exist.

[0167] It should be noted that, in this document, the terms "comprises," "includes," or any other variations thereof are intended to encompass non-exclusive inclusion, such that a process, method, article, or apparatus comprising a series of elements includes not only those elements but also other elements not explicitly listed, or elements inherent to such process, method, article, or apparatus. In the absence of further limitations, an element defined by the phrase "comprising a ..." does not exclude the presence of other identical elements in the process, method, article, or apparatus comprising the element.

[0168] Through the description of the above implementation methods, those skilled in the art can clearly understand that the above-mentioned embodiment methods can be implemented by means of software plus the necessary general hardware platform, and of course can also be implemented by hardware, but in many cases the former is a better implementation method. Based on this understanding, the technical solution of the present application, or the part that contributes to the prior art, can be embodied in the form of a software product, which is stored in a storage medium (such as ROM / RAM, magnetic disk, optical disk), and includes a number of instructions for enabling a terminal (which can be a mobile phone, computer, server, air conditioner, or second terminal device, etc.) to execute the methods of each embodiment of the present application.

[0169] The embodiments of the present application are described above in conjunction with the accompanying drawings, but the present application is not limited to the above-mentioned specific implementation methods. The above-mentioned specific implementation methods are merely illustrative and not restrictive. Under the guidance of this application, ordinary technicians in this field can also make many forms without departing from the purpose of this application and the scope of protection of the claims, all of which are within the protection of this application.

Claims

1. A method for network virtualization of computing resources in an intelligent computing center, characterized in that: include: Step S1: abstract computing resources to form a virtual resource pool; Step S2: creating multiple virtual networks in the virtual resource pool, wherein the multiple virtual networks are isolated from each other; Step S3: configuring corresponding network parameters for each virtual network according to the service type of the computing power service carried by each virtual network; Step S4: Adjust the network parameters of each virtual network according to the operation status data of each virtual network.

2. The method according to claim 1, wherein The multiple virtual networks include a virtual local area network (VLAN), and step S2 includes: Step S21: Determine the identification information of the VLAN; Step S22: creating identification information of the VLAN on the switch, and configuring a port group of the switch corresponding to the identification information of the VLAN in the virtual resource pool; Step S23: Create an access control list (ACL) based on the port group of the switch corresponding to the identification information of the VLAN.

3. The method according to claim 1, wherein The multiple virtual networks include a virtual private network VPN, and step S2 includes: Step S24: Determine the VPN protocol type, encryption algorithm, and tunnel parameters. Step S25: deploying a server instance and a client instance in the virtual resource pool based on the protocol type of the VPN; Step S26: establishing a tunnel connection between the server instance and the client instance based on the tunnel parameters; Step S27: Determine the encryption strategy of the tunnel based on the encryption algorithm.

4. The method according to any one of claims 1 to 3, characterized in that The network parameters include network topology parameters, and the network topology parameters include the connection mode between network nodes of the virtual network and the network hierarchical structure; The step S3 comprises: Step S31: Determine the service characteristics and service scale of the computing service carried by each virtual network according to the service type of the computing service carried by the virtual network; Step S32: determining the number of interactions between the network nodes according to the service characteristics; Step S33: determining a connection mode between the network nodes based on the number of interactions between the network nodes, wherein if the number of interactions between the network nodes is greater than a preset number of interactions, the connection mode between the network nodes is a direct connection; and if the number of interactions between the network nodes is less than or equal to the preset number of interactions, the connection mode between the network nodes is an indirect connection; Step S34: Determine the network hierarchy structure based on the business scale, wherein, when the business scale is greater than the preset business scale, the network hierarchy structure is a multi-level network structure; when the business scale is less than or equal to the preset business scale, the network hierarchy structure is a flat network structure.

5. The method according to claim 4, wherein The service characteristics include data traffic size, delay sensitivity, data integrity requirements and service priority, the network parameters include network performance parameters, and the network performance parameters include network bandwidth requirements, network delay targets, network packet loss rate and QoS policies. Step S3 further includes: Step S35: Determine the network bandwidth requirement based on the data traffic size; Step S36: Determine the network delay target based on the delay sensitivity; Step S37: determining the network packet loss rate based on the data integrity requirement; Step S38: Determine the QoS policy based on the service priority.

6. The method according to claim 1, wherein The step S4 comprises: Step S41: Calculate resource demand data of each virtual network using a pre-trained machine learning model based on the operating status data of each virtual network; Step S42: Adjust the network parameters of each virtual network according to the resource demand data of each virtual network.

7. A computing resource network virtualization device for an intelligent computing center, characterized in that: include: The first processing module is used to abstract computing resources to form a virtual resource pool; A first creation module is used to create multiple virtual networks in the virtual resource pool, where the multiple virtual networks are isolated from each other; A first configuration module is used to configure corresponding network parameters for each virtual network according to the service type of the computing power service carried by each virtual network; The first adjustment module is configured to adjust the network parameters of each virtual network according to the operation status of each virtual network.

8. An electronic device, characterized in that: include: A processor, a memory, and a program stored in the memory and executable on the processor, wherein when the program is executed by the processor, the steps of the method for network virtualization of computing resources of an intelligent computing center as described in any one of claims 1 to 6 are implemented.

9. A computer-readable storage medium, characterized in that The computer-readable storage medium stores a computer program, which, when executed by a processor, implements the steps of the method for network virtualization of computing resources of an intelligent computing center according to any one of claims 1 to 6.

10. A computer program product, characterized in that The method comprises computer instructions, which, when executed by a processor, implement the steps of the method for network virtualization of computing resources of an intelligent computing center as described in any one of claims 1 to 6.

Images