A method, device and readable storage medium for generating computing resource topology

Abstract

The present invention provides a method, device and readable storage medium for generating a computing power resource topology. The method includes: sending a first LLDP message to a neighboring node, wherein the basic field of the data unit LLDPDU in the first LLDP message carries the MAC address information of the network node, and the extended field carries the computing power resource information of the network node; receiving a second LLDP message sent by the neighboring node; sending the MAC address information and computing power resource information corresponding to the network node, as well as the MAC address information of the neighboring node, to a management and control system, so that the management and control system generates a computing power resource topology corresponding to the domain based on the received MAC address information and computing power resource information. The method, device and readable storage medium can solve the problems of the existing method for generating computing power resource topology, which is unable to perceive the computing power resources of the second layer network and the incomplete perception of the computing power resources of the third layer network.

Description

Technical Field

[0001] The present invention relates to the field of network technology, and in particular to a method, device and readable storage medium for generating a computing resource topology. Background Art

[0002] Existing methods for generating computing power resource topology generally use the extended field in the Border Gateway Protocol (BGP) or Internet Protocol version 6 (IPv6) to carry the computing power of the network node itself and send it to adjacent network nodes (i.e., neighboring nodes), thereby realizing the computing power perception of each network node and the generation of computing power resource topology. However, these methods have the problem of being unable to perceive the computing power resources of the second-layer network and incomplete perception of the computing power resources of the third-layer network. Summary of the Invention

[0003] The technical problem to be solved by the present invention is to address the above-mentioned shortcomings of the existing technology and provide a method, device and readable storage medium for generating computing power resource topology, so as to solve the problems that the existing method for generating computing power resource topology cannot perceive the second-layer network computing power resources and has incomplete perception of the third-layer network computing power resources.

[0004] In a first aspect, the present invention provides a method for generating a computing resource topology, which is applied to a network node, and the method includes:

[0005] Sending a first LLDP message to the neighboring node, where a basic field of a data unit LLDPDU in the first LLDP message carries MAC address information of the network node, and an extended field carries computing resource information of the network node;

[0006] Receive a second LLDP message sent by the neighboring node, where a basic field of the LLDPDU in the second LLDP message carries the MAC address information of the neighboring node, and an extended field carries the computing resource information of the neighboring node;

[0007] The MAC address information and computing power resource information corresponding to the network node, as well as the MAC address information of the neighboring node are sent to the management and control system, so that the management and control system generates a computing power resource topology corresponding to the domain based on the received MAC address information and computing power resource information.

[0008] Furthermore, before sending the first LLDP message to the neighboring node, the method further includes:

[0009] Divide computing power resources into general computing power, graphics computing power, and dedicated computing power;

[0010] The computing power resource information is obtained according to the maximum computing power and the remaining computing power of the general computing power, graphics computing power and dedicated computing power corresponding to the network node.

[0011] Furthermore, the computing power resource information of the network node is 128 bits; wherein,

[0012] Bits 1 to 42 are used to indicate the maximum computing power and remaining computing power of the general computing power corresponding to the network node;

[0013] Bits 43 to 84 are used to indicate the maximum computing power and remaining computing power of the graph computing power corresponding to the network node;

[0014] Bits 85 to 126 are used to indicate the maximum computing power and remaining computing power of the dedicated computing power corresponding to the network node;

[0015] Bits 127 to 128 are used to indicate at least one of the following: whether the network node is a computing power node, whether the network node can provide computing power services to the outside world, and whether there is a fault in the computing power resources of the network node.

[0016] Furthermore, the method further comprises:

[0017] Storing the MAC address information and computing power resource information of the network node in the local management information base MIB;

[0018] The MAC address information and computing resource information of the neighbor node are stored in the remote MIB.

[0019] Furthermore, the method further comprises:

[0020] When it is detected that the computing power resource information and / or MAC address information of the neighboring node has changed, the new computing power resource information and / or MAC address information of the neighboring node is sent to the management and control system.

[0021] In a second aspect, the present invention provides a method for generating a computing resource topology, which is applied to a management and control system. The method includes:

[0022] Receive MAC address information and computing resource information corresponding to the network node, as well as MAC address information of neighboring nodes of the network node, sent by each network node;

[0023] The MAC address information and computing power resource information are sent by the network node after sending a first LLDP message to the neighboring node and receiving a second LLDP message sent by the neighboring node. The basic field of the data unit LLDPDU in the first LLDP message carries the MAC address information of the network node, and the extended field carries the computing power resource information of the network node. The basic field of the LLDPDU in the second LLDP message carries the MAC address information of the neighboring node, and the extended field carries the computing power resource information of the neighboring node;

[0024] Generate the computing resource topology corresponding to this domain based on the received MAC address information and computing resource information.

[0025] Furthermore, the method further comprises:

[0026] Receiving new computing resource information and / or MAC address information of the neighboring node sent by the network node;

[0027] Update the computing power resource topology according to the new computing power resource information and / or MAC address information.

[0028] Furthermore, the method further comprises:

[0029] The computing power resource topology corresponding to the domain is sent to the upper-level management and control system, so that the upper-level management and control system generates the computing power resource topology corresponding to the entire network based on the received computing power resource topology and cross-domain connection data.

[0030] In a third aspect, the present invention provides a device for generating a computing resource topology, which is applied to a network node, and the device includes:

[0031] A first message sending module is configured to send a first LLDP message to a neighboring node, wherein a basic field of a data unit LLDPDU in the first LLDP message carries MAC address information of the network node, and an extended field carries computing resource information of the network node;

[0032] A second message receiving module is configured to receive a second LLDP message sent by the neighboring node, wherein the basic field of the LLDPDU in the second LLDP message carries the MAC address information of the neighboring node, and the extended field carries the computing power resource information of the neighboring node;

[0033] The computing power information sending module is connected to the first message sending module and the second message receiving module, and is used to send the MAC address information and computing power resource information corresponding to the network node, as well as the MAC address information of the neighboring node to the management and control system, so that the management and control system generates the computing power resource topology corresponding to the domain based on the received MAC address information and computing power resource information.

[0034] In a fourth aspect, the present invention provides a device for generating a computing resource topology, which is applied to a management and control system, and the device includes:

[0035] A computing power information receiving module is used to receive the MAC address information and computing power resource information corresponding to each network node, as well as the MAC address information of the neighboring nodes of the network node, sent by each network node;

[0036] The MAC address information and computing power resource information are sent by the network node after sending a first LLDP message to the neighboring node and receiving a second LLDP message sent by the neighboring node. The basic field of the data unit LLDPDU in the first LLDP message carries the MAC address information of the network node, and the extended field carries the computing power resource information of the network node. The basic field of the LLDPDU in the second LLDP message carries the MAC address information of the neighboring node, and the extended field carries the computing power resource information of the neighboring node;

[0037] The computing power resource topology generation module is connected to the computing power information receiving module and is used to generate the computing power resource topology corresponding to the domain based on the received MAC address information and computing power resource information.

[0038] In a fifth aspect, the present invention provides a device for generating a computing power resource topology, comprising a memory and a processor, wherein the memory stores a computer program, and the processor is configured to run the computer program to implement the method for generating a computing power resource topology described in the first or second aspect above.

[0039] In a sixth aspect, the present invention provides a computer-readable storage medium having a computer program stored thereon. When the computer program is executed by a processor, the method for generating a computing power resource topology as described in the first or second aspect above is implemented.

[0040] The method, device and readable storage medium for generating computing power resource topology provided by the present invention first send a first LLDP message to a neighboring node, wherein the basic field of the data unit LLDPDU in the first LLDP message carries the MAC address information of the network node, and the extended field carries the computing power resource information of the network node; then, a second LLDP message sent by the neighboring node is received, wherein the basic field of the LLDPDU in the second LLDP message carries the MAC address information of the neighboring node, and the extended field carries the computing power resource information of the neighboring node; then, the MAC address information and computing power resource information corresponding to the network node, as well as the MAC address information of the neighboring node, are sent to the management and control system, so that the management and control system generates the computing power resource topology corresponding to the domain based on the received MAC address information and computing power resource information. By carrying computing power resource information in the extended field of the Layer 2 network Link Layer Discovery Protocol (LLDP), the present invention can realize the interaction of computing power resource information between Layer 2 network nodes and automatically generate computing power resource topology, thereby improving the computing power resource perception efficiency and the utilization rate of computing power resources in computing power network nodes, laying a solid foundation for computing power business scheduling and computing power resource allocation, and solving the problems of the existing computing power resource topology generation method that is unable to perceive Layer 2 network computing power resources and the incomplete perception of Layer 3 network computing power resources. BRIEF DESCRIPTION OF THE DRAWINGS

[0041] Figure 1 This is a flowchart of a method for generating a computing resource topology according to Example 1 of the present invention;

[0042] Figure 2 A scenario diagram of a method for generating a computing resource topology according to an embodiment of the present invention;

[0043] Figure 3 This is another scenario diagram of a method for generating a computing resource topology according to an embodiment of the present invention;

[0044] Figure 4 This is a flowchart of a method for generating a computing resource topology according to Example 2 of the present invention;

[0045] Figure 5 This is a schematic diagram of the structure of a device for generating a computing resource topology according to Example 3 of the present invention;

[0046] Figure 6 This is a schematic diagram of the structure of a device for generating a computing resource topology according to Example 4 of the present invention;

[0047] Figure 7 This is a structural diagram of a device for generating a computing resource topology according to Example 5 of the present invention. DETAILED DESCRIPTION

[0048] In order to enable those skilled in the art to better understand the technical solutions of the present invention, the embodiments of the present invention will be described in further detail below with reference to the accompanying drawings.

[0049] It should be understood that the specific embodiments and drawings described herein are only used to explain the present invention rather than to limit the present invention.

[0050] It is understood that, in the absence of conflict, the various embodiments of the present invention and the various features in the embodiments may be combined with each other.

[0051] It can be understood that, for the convenience of description, the drawings of the present invention only show parts related to the present invention, while parts unrelated to the present invention are not shown in the drawings.

[0052] It can be understood that each unit and module involved in the embodiments of the present invention may correspond to only one physical structure, or may be composed of multiple physical structures, or multiple units and modules may be integrated into one physical structure.

[0053] It can be understood that the terms "first", "second", etc. in the embodiments of the present invention are used to distinguish different objects, or to distinguish different processing of the same object, rather than to describe a specific order of objects.

[0054] It will be understood that, without conflict, the functions and steps marked in the flowcharts and block diagrams of the present invention may occur in an order different from that marked in the drawings.

[0055] It is understood that the flowcharts and block diagrams of the present invention illustrate the possible architectures, functions, and operations of the systems, devices, equipment, and methods according to various embodiments of the present invention. Each box in the flowchart or block diagram may represent a unit, module, program segment, or code, which contains executable instructions for implementing the specified functions. Moreover, each box or combination of boxes in the block diagram and flowchart may be implemented using a hardware-based system that implements the specified functions, or may be implemented using a combination of hardware and computer instructions.

[0056] It can be understood that the units and modules involved in the embodiments of the present invention can be implemented by software or hardware. For example, the units and modules can be located in a processor.

[0057] Application Overview

[0058] With the development of the digital economy, computing power has become a new productive force in the digital economy of the new era. By transforming computing power into infrastructure, developing it into a computing network, reducing computing service costs, and enabling computing power to serve all aspects of the digital society, we can promote more robust and high-quality development of the digital economy. However, how to automatically sense the computing power of each node in the network and generate a computing resource topology has become a pressing issue.

[0059] The existing method for generating computing power resource topology generally uses the extended field in the Border Gateway Protocol (BGP) or Internet Protocol version 6 (IPv6) to carry the computing power of the network node itself and send it to the adjacent network nodes (i.e., neighboring nodes), thereby realizing the computing power perception of each network node and the generation of computing power resource topology.

[0060] Specifically, the network node sends the IPv6 data packet to other node devices within the same AS (Autonomous System) by inserting the node's own computing resources in the IPv6 extension field. Similarly, the network node can also obtain IPv6 data packets from other node devices in the same AS and record their computing resources. This enables the automatic discovery of the computing resource topology of the three-layer network nodes that support IPv6 in the same AS. However, this method can only automatically obtain the computing resource topology that supports the IPv6 protocol within the same AS. The interaction of computing resources between different ASs is achieved through the BGP protocol. The three-layer network device at the edge of the AS transmits all the computing resources perceived under its own AS to the adjacent AS network through the BGP protocol, thereby realizing the cross-AS transmission of computing resources. However, since BGP and IPv6 are both Layer 3 (transport layer) network protocols and can only be transmitted through Layer 3 networks, and there are also a large number of Layer 2 network nodes in the existing network that also have computing resources, existing methods are unable to perceive the computing resources of Layer 2 (data link layer) network nodes and can only generate computing resource topology through manual reporting. However, manual reporting methods cannot perceive the computing resources of the current network in real time and have a serious lag, which has a serious impact on the allocation of computing network services and the scheduling of computing resources. In addition, many devices in the existing network only support the IPv4 protocol and do not support the IPv6 protocol; and the BGP protocol is an external gateway protocol used between different ASes and is not applicable to network nodes within the same autonomous system (AS). Therefore, existing methods can only obtain the computing resource topology of Layer 3 network nodes that support IPv6 or between different ASes, and cannot achieve automatic discovery of the computing resource topology of Layer 2 network nodes.

[0061] In response to the above technical problems, the concept of this application is to provide a method, device and readable storage medium for generating computing power resource topology. By carrying computing power resource information in the extended field of the Layer 2 network Link Layer Discovery Protocol (LLDP), it is possible to realize the interaction of computing power resource information between Layer 2 network nodes and automatically generate computing power resource topology, thereby improving the computing power resource perception efficiency and the utilization rate of computing power resources in computing power network nodes, laying a solid foundation for computing power business scheduling and computing power resource allocation, and solving the industry's problems of inability to perceive Layer 2 network computing power resources and incomplete perception of Layer 3 network computing power resources.

[0062] After introducing the basic principles of the present application, various non-limiting embodiments of the present application will be described in detail with reference to the accompanying drawings.

[0063] Example 1:

[0064] This embodiment provides a method for generating a computing resource topology, which is applied to network nodes, such as Figure 1 As shown, the method includes:

[0065] Step S101: Send a first LLDP message to a neighboring node, wherein the basic field of the data unit LLDPDU in the first LLDP message carries the MAC (Media Access Control) address information of the network node, and the extended field carries the computing resource information of the network node.

[0066] In this embodiment, the network node can be any network device, such as a switch, router, server, or PC, or any combination thereof. The network node enables the LLDP protocol in advance and defines the LLDPDU extension field. The network node sends a first LLDP message to a neighboring node, allowing the neighboring node to obtain the network node's MAC address and computing resource information.

[0067] Optionally, before sending the first LLDP message to the neighboring node, the method further includes:

[0068] Divide computing power resources into general computing power, graphics computing power, and dedicated computing power;

[0069] The computing power resource information is obtained according to the maximum computing power and the remaining computing power of the general computing power, graphics computing power and dedicated computing power corresponding to the network node.

[0070] In this embodiment, computing power resources can be divided into three categories: general computing power (CPU), graphics computing power (GPU), and dedicated computing power. General floating-point computing services mainly rely on the CPU for calculations, which require general computing power; image processing services mainly rely on the GPU for calculations, which require graphics computing power; some special computing power services may use specialized processors (such as FPGAs) and require dedicated computing power. Before sending the first LLDP message to the neighboring node, the network node inserts the maximum computing power and the remaining computing power of the three computing powers into the extended field of the LLDP protocol data unit (LLDPDU), and discovers and announces the computing power resource information of the node devices in the network through the LLDP protocol.

[0071] Optionally, the computing power resource information of the network node is 128 bits; wherein,

[0072] Bits 1 to 42 are used to indicate the maximum computing power and remaining computing power of the general computing power corresponding to the network node;

[0073] Bits 43 to 84 are used to indicate the maximum computing power and remaining computing power of the graph computing power corresponding to the network node;

[0074] Bits 85 to 126 are used to indicate the maximum computing power and remaining computing power of the dedicated computing power corresponding to the network node;

[0075] Bits 127 to 128 are used to indicate at least one of the following: whether the network node is a computing power node, whether the network node can provide computing power services to the outside world, and whether there is a fault in the computing power resources of the network node.

[0076] In this embodiment, the extended field of the LLDPDU occupies a total of 128 bits to store 128 bits of computing resource information. The specific definition can be as follows:

[0077] Bits 1 to 42 are used for general computing power (CPU), where bits 1 to 21 represent the maximum general computing power of the network node, and bits 22 to 42 represent the remaining general computing power of the network node.

[0078] Bits 43 to 84 are used for graphics computing power (GPU), where bits 43 to 63 represent the maximum graphics computing power of the network node, and bits 64 to 84 represent the remaining graphics computing power of the network node.

[0079] Bits 85 to 126 are used for dedicated computing power, where bits 85 to 105 represent the maximum dedicated computing power of the network node, and bits 106 to 126 represent the remaining dedicated computing power of the network node.

[0080] Bits 127 and 128 are used to indicate at least one of the following: whether the network node is a computing power node, whether the network node can provide computing power services to the outside world, and whether the computing power resources of the network node are faulty. For example, 00 indicates that the network node is not a computing power node and can only forward computing power data in the computing power network but cannot process computing power data; 01 indicates that the network node is a computing power node but currently does not provide external computing power services, that is, the computing power of the node is not currently intended to be provided to the computing power network (it may be reserved for internal use); 10 indicates that the network node is a computing power node and can normally provide computing power services to the outside world; 11 indicates that some computing power resources of the network node are faulty and need to be repaired as soon as possible (for example, the GPU computing power is faulty, but the CPU computing power is normal. Although the GPU maximum and remaining computing power are both 0, they can be restored to normal through repair and need to be repaired as soon as possible. This is different from the network node that does not provide GPU computing power and the GPU maximum and remaining computing power are both 0).

[0081] It should be noted that the current network includes both network nodes with computing resources and network nodes without computing resources, which are distinguished by the 127th and 128th bits to facilitate the establishment of the computing resource topology.

[0082] It should be noted that a network node can be composed of one or more network devices. When calculating computing power resources, the computing power resources of the network node are equal to the sum of the resources of all network devices within the node. The MAC address of a network node can be represented by the MAC address of the egress device (or core device) in the node. A node has only one unique MAC address. The present invention does not strictly distinguish between network "nodes" and network "devices" in its description.

[0083] Step S102: Receive a second LLDP message sent by the neighboring node, where the basic field of the LLDPDU in the second LLDP message carries the MAC address information of the neighboring node, and the extended field carries the computing resource information of the neighboring node.

[0084] In this embodiment, the neighboring node can be any network device, such as a switch, router, server, or PC, or any combination thereof. The neighboring node has previously enabled the LLDP protocol and defined the LLDPDU extension field. By receiving the second LLDP message from the neighboring node, the network node can obtain the neighboring node's MAC address and computing resource information. The computing resource information of the neighboring node is defined similarly to that of the network node.

[0085] Optionally, the method further includes:

[0086] Storing the MAC address information and computing power resource information of the network node in the local management information base MIB;

[0087] The MAC address information and computing resource information of the neighbor node are stored in the remote MIB.

[0088] In this embodiment, each network node has two management information bases, one for maintaining the MIB information of the local node, and the other for maintaining the MIB information of the remote node. The network node adds its own computing resource information (128 bits defined in the LLDPDU extension field) and MAC address information to the local MIB, and stores the computing resource information and MAC address information of the neighboring node in the remote MIB.

[0089] Step S103: Send the MAC address information and computing power resource information corresponding to the network node, as well as the MAC address information of the neighboring node, to the management and control system, so that the management and control system generates a computing power resource topology corresponding to the domain based on the received MAC address information and computing power resource information.

[0090] In this embodiment, each control system corresponds to an AS domain. The control system generates the computing resource topology corresponding to the domain by receiving local information and neighbor information from each network node within the domain. It should be noted that network nodes can also send computing resource information from their neighbor nodes to the control system. After receiving the MAC address information and computing resource information from each network node and its neighbor nodes, the control system performs a computing resource topology analysis to generate the computing resource topology corresponding to the domain.

[0091] In this embodiment, since there are still nodes in the current network that do not have computing power resources, and these nodes are connected to the nodes with computing power resources, the management and control system can discover all network topologies by receiving local MAC address information and computing power resource information sent by each network node in the domain and the MAC address information of the neighboring nodes, and the corresponding computing power resource topology can be obtained by excluding the non-computing power nodes.

[0092] In this embodiment, the computing power resource topology includes the network connection relationship and computing power resource information of the second layer and above (not only the second layer network, but also the network connection relationship and computing power resource information of all second layer network nodes and all third layer network nodes since the third layer network supports the second layer protocol), thereby solving the problem of incomplete perception of the third layer network computing power resources in the existing technology.

[0093] In this embodiment, the management and control system corresponding to each domain can send the computing power resource topology corresponding to the domain to the upper-level management and control system, so that the upper-level management and control system can generate the computing power resource topology corresponding to the entire network based on the received computing power resource topology and cross-domain connection data.

[0094] Optionally, the method further includes:

[0095] When it is detected that the computing power resource information and / or MAC address information of the neighboring node has changed, the new computing power resource information and / or MAC address information of the neighboring node is sent to the management and control system.

[0096] In this embodiment, when a network node discovers that the computing power resource information and / or MAC address information of a neighboring node has changed, it notifies the management and control system (network management), and the management and control system updates the computing power resource topology based on the new computing power resource information and / or MAC address information.

[0097] In a specific embodiment, Figure 2 As shown in the figure, taking a high-performance switch with computing resources as an example, Switch-A and Switch-B use LLDP to discover neighboring nodes and their computing resources as follows:

[0098] 1. Switch-A extracts the computing resource information and MAC address from the local MIB, uses the LLDPDU basic field to carry the MAC address information, and the extended field to carry the computing resource information, and sends the local computing resource information and MAC address information to Switch-B.

[0099] 2. Switch-B extracts the computing resource information and MAC address information about Switch-A from the received LLDP message and stores it in the remote MIB of the local device so that the management system (network management) can extract the computing resource topology information.

[0100] 3. Similarly, Switch-B sends its computing resource information and MAC address information to Switch-A. Switch-A extracts the computing resource information and MAC address information about Switch-B from the received LLDP message and stores it in the remote MIB of the local device so that the management system (network management) can extract computing resource topology information.

[0101] 4. Switch-A and Switch-B send their own computing resource information and MAC address information, as well as each other's MAC address information or their MAC address information and computing resource information to the management and control system through the southbound interface. The management and control system extracts local information and neighbor information and performs computing resource topology analysis to ultimately determine the computing resource topology of the local domain.

[0102] 5. After Switch-A and Switch-B discover changes in their neighbor's computing resource information and MAC address information, they notify the management and control system. The management and control system extracts local information and neighbor information from Switch-A and Switch-B through the southbound interface and performs computing resource topology analysis to update the computing resource topology.

[0103] In another specific embodiment, Figure 3 As shown in the figure, the target network includes two AS domains, A and B. Each domain contains multiple network nodes. The method for discovering the computing resource topology through LLDP is as follows:

[0104] 1. All network nodes in the same AS domain enable the LLDP protocol and define the LLDPDU extended field. All network nodes in the same AS domain send LLDP packets to their neighboring nodes.

[0105] 2. After LLDP interaction between network nodes becomes stable, each domain's management and control system obtains the computing resource topology corresponding to each network node in the corresponding domain through southbound interface polling or node active reporting.

[0106] 3. LLDP data collection between different AS domains: The upper-layer management and control system obtains the computing resource topology of each domain (domain A and domain B) based on the northbound interface of each domain management and control system.

[0107] 4. Generation of the entire network computing power resource topology: The upper-level management and control system uses the collected single-domain computing power resource topology and cross-domain connection data to splice the computing power resource topology of each domain and automatically generate the entire network computing power resource topology.

[0108] The method for generating a computing power resource topology provided by an embodiment of the present invention first sends a first LLDP message to a neighboring node, wherein the basic field of the data unit LLDPDU in the first LLDP message carries the MAC address information of the network node, and the extended field carries the computing power resource information of the network node; then, a second LLDP message sent by the neighboring node is received, wherein the basic field of the LLDPDU in the second LLDP message carries the MAC address information of the neighboring node, and the extended field carries the computing power resource information of the neighboring node; then, the MAC address information and computing power resource information corresponding to the network node, as well as the MAC address information of the neighboring node, are sent to a management and control system, so that the management and control system generates a computing power resource topology corresponding to the domain based on the received MAC address information and computing power resource information. By carrying computing power resource information in the extended field of the Layer 2 network Link Layer Discovery Protocol (LLDP), the present invention can realize the interaction of computing power resource information between Layer 2 network nodes and automatically generate computing power resource topology, thereby improving the computing power resource perception efficiency and the utilization rate of computing power resources in computing power network nodes, laying a solid foundation for computing power business scheduling and computing power resource allocation, and solving the problems of the existing computing power resource topology generation method that is unable to perceive Layer 2 network computing power resources and the incomplete perception of Layer 3 network computing power resources.

[0109] Example 2:

[0110] like Figure 4 As shown, this embodiment provides a method for generating a computing resource topology, which is applied to a management and control system. The method includes:

[0111] Step S201: receiving MAC address information and computing resource information corresponding to each network node, as well as MAC address information of neighboring nodes of the network node, sent by each network node;

[0112] Among them, the MAC address information and computing power resource information are sent by the network node after sending a first LLDP message to the neighbor node and receiving a second LLDP message sent by the neighbor node. The basic field of the data unit LLDPDU in the first LLDP message carries the MAC address information of the network node, and the extended field carries the computing power resource information of the network node. The basic field of the LLDPDU in the second LLDP message carries the MAC address information of the neighbor node, and the extended field carries the computing power resource information of the neighbor node.

[0113] In this embodiment, computing resources can be divided into three categories: general computing power (CPU), graphics computing power (GPU), and dedicated computing power. General floating-point computing services primarily rely on the CPU and require general computing power; image processing services primarily rely on the GPU and require graphics computing power; and some specialized computing services may utilize specialized processors (such as FPGAs) and require dedicated computing power.

[0114] In this embodiment, the computing power resource information is 128 bits;

[0115] Bits 1 to 42 are used to indicate the maximum and remaining computing power of the general computing power;

[0116] Bits 43 to 84 are used to indicate the maximum and remaining computing power of the graphics computing power;

[0117] Bits 85 to 126 are used to indicate the maximum and remaining computing power of the dedicated computing power;

[0118] Bits 127 to 128 are used to indicate at least one of the following: whether the network node is a computing power node, whether the network node can provide computing power services to the outside world, and whether there is a fault in the computing power resources of the network node.

[0119] Step S202: Generate a computing resource topology corresponding to the domain based on the received MAC address information and computing resource information.

[0120] In this embodiment, the management and control system can generate a computing resource topology corresponding to the domain by receiving local information and neighbor information sent by each network node in the domain.

[0121] Optionally, the method further includes:

[0122] Receiving new computing resource information and / or MAC address information of the neighboring node sent by the network node;

[0123] Update the computing power resource topology according to the new computing power resource information and / or MAC address information.

[0124] In this embodiment, when a network node discovers that the computing power resource information and / or MAC address information of a neighboring node has changed, it notifies the management and control system, and the management and control system updates the computing power resource topology based on the new computing power resource information and / or MAC address information.

[0125] Optionally, the method further includes:

[0126] The computing power resource topology corresponding to the domain is sent to the upper-level management and control system, so that the upper-level management and control system generates the computing power resource topology corresponding to the entire network based on the received computing power resource topology and cross-domain connection data.

[0127] In this embodiment, the management and control system corresponding to each domain can send the computing power resource topology corresponding to the domain to the upper-level management and control system. The upper-level management and control system uses the collected single-domain computing power resource topology and cross-domain connection data to splice the computing power resource topology of each domain to complete the automatic generation of the computing power resource topology of the entire network.

[0128] Example 3:

[0129] like Figure 5 As shown, this embodiment provides a device for generating a computing resource topology, which is applied to a network node and is used to execute the method for generating a computing resource topology in the above embodiment 1. The device includes:

[0130] A first message sending module 11 is configured to send a first LLDP message to a neighboring node, wherein the basic field of the data unit LLDPDU in the first LLDP message carries the MAC address information of the network node, and the extended field carries the computing resource information of the network node;

[0131] A second message receiving module 12 is configured to receive a second LLDP message sent by the neighboring node, wherein the basic field of the LLDPDU in the second LLDP message carries the MAC address information of the neighboring node, and the extended field carries the computing resource information of the neighboring node;

[0132] The computing power information sending module 13 is connected to the first message sending module 11 and the second message receiving module 12, and is used to send the MAC address information and computing power resource information corresponding to the network node, as well as the MAC address information of the neighboring node to the management and control system, so that the management and control system generates the computing power resource topology corresponding to the domain based on the received MAC address information and computing power resource information.

[0133] Optionally, the device further comprises:

[0134] The computing power division module is used to divide computing power resources into general computing power, graphics computing power, and dedicated computing power;

[0135] The computing power information generation module is used to obtain the computing power resource information based on the maximum computing power and remaining computing power of the general computing power, graphic computing power and dedicated computing power corresponding to the network node.

[0136] Optionally, the computing power resource information of the network node is 128 bits; wherein,

[0137] Bits 1 to 42 are used to indicate the maximum computing power and remaining computing power of the general computing power corresponding to the network node;

[0138] Bits 43 to 84 are used to indicate the maximum computing power and remaining computing power of the graph computing power corresponding to the network node;

[0139] Bits 85 to 126 are used to indicate the maximum computing power and remaining computing power of the dedicated computing power corresponding to the network node;

[0140] Bits 127 to 128 are used to indicate at least one of the following: whether the network node is a computing power node, whether the network node can provide computing power services to the outside world, and whether there is a fault in the computing power resources of the network node.

[0141] Optionally, the device further comprises:

[0142] A first storage module is used to store the MAC address information and computing resource information of the network node in a local management information base MIB;

[0143] The second storage module is used to store the MAC address information and computing resource information of the neighbor node in the remote MIB.

[0144] Optionally, the device further comprises:

[0145] The update information sending module is used to send the new computing power resource information and / or MAC address information of the neighboring node to the management and control system when it is detected that the computing power resource information and / or MAC address information of the neighboring node has changed.

[0146] Example 4:

[0147] like Figure 6 As shown, this embodiment provides a device for generating a computing resource topology, which is applied to a management and control system and is used to execute the method for generating a computing resource topology in the above embodiment 2. The device includes:

[0148] The computing power information receiving module 21 is used to receive the MAC address information and computing power resource information corresponding to each network node, as well as the MAC address information of the neighboring nodes of the network node, sent by each network node;

[0149] The MAC address information and computing power resource information are sent by the network node after sending a first LLDP message to the neighboring node and receiving a second LLDP message sent by the neighboring node. The basic field of the data unit LLDPDU in the first LLDP message carries the MAC address information of the network node, and the extended field carries the computing power resource information of the network node. The basic field of the LLDPDU in the second LLDP message carries the MAC address information of the neighboring node, and the extended field carries the computing power resource information of the neighboring node;

[0150] The computing power resource topology generating module 22 is connected to the computing power information receiving module 21 and is used to generate a computing power resource topology corresponding to the domain according to the received MAC address information and computing power resource information.

[0151] Optionally, the device further comprises:

[0152] An update information receiving module, configured to receive new computing resource information and / or MAC address information of the neighboring node sent by the network node;

[0153] A computing power resource topology update module is used to update the computing power resource topology according to the new computing power resource information and / or MAC address information.

[0154] Optionally, the device further comprises:

[0155] The computing power resource topology sending module is used to send the computing power resource topology corresponding to the domain to the upper-level management and control system, so that the upper-level management and control system generates the computing power resource topology corresponding to the entire network based on the received computing power resource topology and cross-domain connection data.

[0156] Example 5:

[0157] refer to Figure 7 This embodiment provides a device for generating a computing power resource topology, including a memory 31 and a processor 32. The memory 31 stores a computer program, and the processor 32 is configured to run the computer program to execute the method for generating a computing power resource topology in Example 1.

[0158] The memory 31 is connected to the processor 32 . The memory 31 may be a flash memory, a read-only memory, or other memory. The processor 32 may be a central processing unit or a single-chip microcomputer.

[0159] Example 6:

[0160] This embodiment provides a computer-readable storage medium, on which a computer program is stored. When the computer program is executed by a processor, the method for generating a computing power resource topology in the above-mentioned embodiment 1 is implemented.

[0161] The computer-readable storage medium includes volatile or non-volatile, removable or non-removable media implemented in any method or technology for storing information (such as computer-readable instructions, data structures, computer program modules or other data). Computer-readable storage media include, but are not limited to, RAM (Random Access Memory), ROM (Read-Only Memory), EEPROM (Electrically Erasable Programmable read only memory), flash memory or other memory technology, CD-ROM (Compact Disc Read-Only Memory), digital versatile disks (DVD) or other optical disk storage, magnetic cassettes, magnetic tape, magnetic 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 a computer.

[0162] To summarize, the method, device, and readable storage medium for generating a computing power resource topology provided by an embodiment of the present invention first send a first LLDP message to a neighboring node, wherein the basic field of the data unit LLDPDU in the first LLDP message carries the MAC address information of the network node, and the extended field carries the computing power resource information of the network node; then, a second LLDP message sent by the neighboring node is received, wherein the basic field of the LLDPDU in the second LLDP message carries the MAC address information of the neighboring node, and the extended field carries the computing power resource information of the neighboring node; then, the MAC address information and computing power resource information corresponding to the network node, as well as the MAC address information of the neighboring node are sent to the management and control system, so that the management and control system generates a computing power resource topology corresponding to the domain based on the received MAC address information and computing power resource information. By carrying computing power resource information in the extended field of the Layer 2 network Link Layer Discovery Protocol (LLDP), the present invention can realize the interaction of computing power resource information between Layer 2 network nodes and automatically generate computing power resource topology, thereby improving the computing power resource perception efficiency and the utilization rate of computing power resources in computing power network nodes, laying a solid foundation for computing power business scheduling and computing power resource allocation, and solving the problems of the existing computing power resource topology generation method that is unable to perceive Layer 2 network computing power resources and the incomplete perception of Layer 3 network computing power resources.

[0163] It will be understood that the above embodiments are merely exemplary embodiments for illustrating the principles of the present invention, and the present invention is not limited thereto. Those skilled in the art will appreciate that various modifications and improvements can be made without departing from the spirit and substance of the present invention, and such modifications and improvements are also considered to be within the scope of protection of the present invention.

Claims

1. A method for generating computing resource topology, characterized in that: Applied to a network node, the method includes: Sending a first LLDP message to the neighboring node, where a basic field of a data unit LLDPDU in the first LLDP message carries MAC address information of the network node, and an extended field carries computing resource information of the network node; Receive a second LLDP message sent by the neighboring node, where a basic field of the LLDPDU in the second LLDP message carries the MAC address information of the neighboring node, and an extended field carries the computing resource information of the neighboring node; The MAC address information and computing power resource information corresponding to the network node, as well as the MAC address information of the neighboring node are sent to the management and control system, so that the management and control system generates a computing power resource topology corresponding to the domain based on the received MAC address information and computing power resource information.

2. The method according to claim 1, characterized in that Before sending the first LLDP message to the neighboring node, the method further includes: Divide computing power resources into general computing power, graphics computing power, and dedicated computing power; The computing power resource information is obtained according to the maximum computing power and the remaining computing power of the general computing power, graphics computing power and dedicated computing power corresponding to the network node.

3. The method according to claim 2, characterized in that The computing power resource information of the network node is 128 bits; wherein, Bits 1 to 42 are used to indicate the maximum computing power and remaining computing power of the general computing power corresponding to the network node; Bits 43 to 84 are used to indicate the maximum computing power and remaining computing power of the graph computing power corresponding to the network node; Bits 85 to 126 are used to indicate the maximum computing power and remaining computing power of the dedicated computing power corresponding to the network node; Bits 127 to 128 are used to indicate at least one of the following: whether the network node is a computing power node, whether the network node can provide computing power services to the outside world, and whether there is a fault in the computing power resources of the network node.

4. The method according to claim 1, wherein The method further comprises: Storing the MAC address information and computing power resource information of the network node in the local management information base MIB; The MAC address information and computing resource information of the neighbor node are stored in the remote MIB.

5. The method according to claim 1, wherein The method further comprises: When it is detected that the computing power resource information and / or MAC address information of the neighboring node has changed, the new computing power resource information and / or MAC address information of the neighboring node is sent to the management and control system.

6. A method for generating computing resource topology, characterized in that: Applied to a management and control system, the method includes: Receive MAC address information and computing resource information corresponding to the network node, as well as MAC address information of neighboring nodes of the network node, sent by each network node; The MAC address information and computing power resource information are sent by the network node after sending a first LLDP message to the neighboring node and receiving a second LLDP message sent by the neighboring node. The basic field of the data unit LLDPDU in the first LLDP message carries the MAC address information of the network node, and the extended field carries the computing power resource information of the network node. The basic field of the LLDPDU in the second LLDP message carries the MAC address information of the neighboring node, and the extended field carries the computing power resource information of the neighboring node; Generate the computing resource topology corresponding to this domain based on the received MAC address information and computing resource information.

7. The method according to claim 6, characterized in that The method further comprises: Receiving new computing resource information and / or MAC address information of the neighboring node sent by the network node; Update the computing power resource topology according to the new computing power resource information and / or MAC address information.

8. The method according to claim 6, characterized in that The method further comprises: The computing power resource topology corresponding to the domain is sent to the upper-level management and control system, so that the upper-level management and control system generates the computing power resource topology corresponding to the entire network based on the received computing power resource topology and cross-domain connection data.

9. A device for generating computing resource topology, characterized in that: Applied to a network node, the device includes: A first message sending module is configured to send a first LLDP message to a neighboring node, wherein a basic field of a data unit LLDPDU in the first LLDP message carries MAC address information of the network node, and an extended field carries computing resource information of the network node; A second message receiving module is configured to receive a second LLDP message sent by the neighboring node, wherein the basic field of the LLDPDU in the second LLDP message carries the MAC address information of the neighboring node, and the extended field carries the computing power resource information of the neighboring node; The computing power information sending module is connected to the first message sending module and the second message receiving module, and is used to send the MAC address information and computing power resource information corresponding to the network node, as well as the MAC address information of the neighboring node to the management and control system, so that the management and control system generates the computing power resource topology corresponding to the domain based on the received MAC address information and computing power resource information.

10. A device for generating computing resource topology, characterized in that: Applied to a management and control system, the device includes: A computing power information receiving module is used to receive the MAC address information and computing power resource information corresponding to each network node, as well as the MAC address information of the neighboring nodes of the network node, sent by each network node; The MAC address information and computing power resource information are sent by the network node after sending a first LLDP message to the neighboring node and receiving a second LLDP message sent by the neighboring node. The basic field of the data unit LLDPDU in the first LLDP message carries the MAC address information of the network node, and the extended field carries the computing power resource information of the network node. The basic field of the LLDPDU in the second LLDP message carries the MAC address information of the neighboring node, and the extended field carries the computing power resource information of the neighboring node; The computing power resource topology generation module is connected to the computing power information receiving module and is used to generate the computing power resource topology corresponding to the domain based on the received MAC address information and computing power resource information.

11. A device for generating computing resource topology, characterized in that: It includes a memory and a processor, wherein the memory stores a computer program, and the processor is configured to run the computer program to implement the method for generating a computing power resource topology as described in any one of claims 1 to 5, or to implement the method for generating a computing power resource topology as described in any one of claims 6 to 8.

12. 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 method for generating a computing power resource topology as described in any one of claims 1-5, or implements the method for generating a computing power resource topology as described in any one of claims 6-8.

Images