Computing power routing method, electronic device and computer program product

Abstract

The invention discloses a computing power routing method, electronic equipment and a computer program product. The method relates to the technical field of computer networks, and comprises the following steps: receiving a computing power service request sent by a user side; based on the computing power service request, determining a target computing power service node with the highest computing power factor from a target computing power routing table and a routing path from the user side to the target computing power service node; wherein a corresponding relationship between the computing power factor and the computing power service node and a routing path leading to the corresponding computing power service node are stored in the target computing power routing table; the computing power factor is used for measuring the computing power service quality of the corresponding computing power service node. According to the invention, the technical problems of inaccurate calculation power service node selection and path optimization and low efficiency in the prior art are solved.

Description

Technical Field

[0001] The present invention relates to the technical field of computer networks, and in particular, to a computing power routing method, an electronic device, and a computer program product. Background Art

[0002] With the continuous development of Internet technology, users' access requirements for computing resources are constantly increasing. In particular, the requirements for computing resources in businesses such as high-performance computing, big data processing, and cloud-native computing have shown explosive growth. Computing power routing is one of the core technologies of the control plane of the computing power network. Through this technology, the computing power network can perceive information such as users' service requirements, computing power resources, and network resources, realize the network distribution of computing power service node resource information, and finally dynamically select the forwarding path of service traffic according to service policies. In the related art, static scheduling and path selection strategies are mainly adopted. For example, the resource scheduling priority is determined according to the resource usage of nodes. This static scheduling strategy cannot reflect the changes in the computing power of computing power service nodes in real time, and it is also difficult to cope with sudden computing power requirements, which easily leads to problems such as low utilization efficiency of computing resources, inaccurate selection of computing power service nodes, and path optimization.

[0003] In response to the above problems, no effective solution has been proposed yet. Summary of the Invention

[0004] Embodiments of the present invention provide a computing power routing method, an electronic device, and a computer program product to at least solve the technical problems of inaccurate and inefficient selection of computing power service nodes and path optimization in the related art.

[0005] According to one aspect of the embodiments of the present invention, a computing power routing method is provided, including: receiving a computing power service request sent by a user terminal; based on the computing power service request, determining a target computing power service node with the highest computing power factor from a target computing power routing table, and a routing path from the user terminal to the target computing power service node; wherein, the target computing power routing table stores the corresponding relationship between the computing power factor and the computing power service node, and the routing path to the corresponding computing power service node; the computing power factor is used to measure the computing power service quality of the corresponding computing power service node.

[0006] According to another aspect of the embodiments of the present invention, an electronic device is further provided, including one or more processors and a memory, where the memory is used to store one or more programs, and when the one or more programs are executed by the one or more processors, the one or more processors are enabled to implement any one of the computing power routing methods.

[0007] According to another aspect of the embodiments of the present invention, there is also provided a computer program product, including a computer program which, when executed by a processor, implements the steps of any one of the above-mentioned computing power routing methods.

[0008] In the embodiments of the present invention, by receiving a computing power service request sent by a user terminal; based on the computing power service request, determining a target computing power service node with the highest computing power factor from a target computing power routing table, and a routing path from the user terminal to the target computing power service node; wherein, the target computing power routing table stores the correspondence between the computing power factor and the computing power service node, and the routing path to the corresponding computing power service node; the computing power factor is used to measure the computing power service quality of the corresponding computing power service node, achieving the purpose of constructing and utilizing the correspondence between the computing power factor and the computing power service node in the target computing power routing table, and automatically selecting the node with the highest computing power service quality and its routing path according to the user request, thereby realizing the technical effect of improving the efficiency and accuracy of the selection of the computing power service node and path optimization, and further solving the technical problems of inaccurate selection and low efficiency of the computing power service node and path optimization in the related art. BRIEF DESCRIPTION OF THE DRAWINGS

[0009] The drawings described herein are used to provide a further understanding of the present invention, form a part of this application, and the schematic embodiments of the present invention and their descriptions are used to explain the present invention, and do not constitute an improper limitation to the present invention. In the drawings:

[0010] Figure 1 is a flowchart of a computing power routing method according to an embodiment of the present invention;

[0011] Figure 2 is a schematic diagram of a computing power network deployment networking based on the IS-IS protocol according to an embodiment of the present invention;

[0012] Figure 3 is a schematic diagram of an optional IS-IS protocol structure according to an embodiment of the present invention;

[0013] Figure 4 is a schematic diagram of a process for generating a computing power routing table based on the IS-IS protocol according to an embodiment of the present invention;

[0014] Figure 5 is a schematic diagram of a process for generating a computing power routing table corresponding to a computing power service node under multiple computing power service preferences according to an embodiment of the present invention;

[0015] Figure 6 is a schematic diagram of a process for generating a computing power routing table corresponding to a computing power service node under multiple computing power service preferences according to an embodiment of the present invention;

[0016] Figure 7 It is an optional computing power routing flowchart according to an embodiment of the present invention;

[0017] Figure 8 It is a schematic diagram of an optional IS-IS protocol message interaction process according to an embodiment of the present invention;

[0018] Figure 9 It is a schematic diagram of an optional IS-IS computing power routing table generation process according to an embodiment of the present invention;

[0019] Figure 10 It is a schematic diagram of an optional computing power service scheduling process for different client terminals based on different service requirements according to an embodiment of the present invention;

[0020] Figure 11 It is a schematic diagram of a computing power routing device according to an embodiment of the present invention. Detailed implementation manners

[0021] In order to enable those skilled in the art to better understand the solution of the present invention, the technical solutions in the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings in the embodiments of the present invention. Obviously, the described embodiments are only a part of the embodiments of the present invention, rather than all the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those of ordinary skill in the art without creative efforts shall fall within the protection scope of the present invention.

[0022] It should be noted that the terms "first", "second", etc. in the description and claims of the present invention and the above drawings are used to distinguish similar objects, and do not necessarily need to describe a specific order or sequence. It should be understood that such data can be interchanged under appropriate circumstances so that the embodiments of the present invention described herein can be implemented in an order different from those illustrated or described herein. In addition, the terms "comprising" and "having" and any variations thereof are intended to cover non-exclusive inclusion. For example, a process, method, system, product or device that includes a series of steps or units does not necessarily have to be limited to those steps or units clearly listed, but may include other steps or units not clearly listed or inherent to these processes, methods, products or devices.

[0023] According to an embodiment of the present invention, a method embodiment of computing power routing is provided. It should be noted that the steps shown in the flowchart of the accompanying drawings can be executed in a computer system such as a set of computer executable instructions, and although the logical order is shown in the flowchart, in some cases, the steps shown or described can be executed in an order different from that here.

[0024] Figure 1is a flowchart of a computing power routing method according to an embodiment of the present invention, as Figure 1 shown. The method includes the following steps:

[0025] Step S102, receiving a computing power service request sent by a user terminal;

[0026] Optionally, the methods of steps S102 to S104 in this embodiment can be applied to a computing power network based on the IS-IS protocol. Figure 2 is an optional deployment networking schematic diagram of a computing power network based on the IS-IS protocol according to an embodiment of the present invention, as Figure 2 shown. The computing power network includes a computing power service scheduling platform, multiple computing power service nodes, and multiple computing power gateways. Among them, the multiple computing power gateways in the computing power network include a user-side computing power gateway and a computing power resource-side computing power gateway. The execution subject corresponding to steps S102 to S104 can be the user-side gateway. The computing power service nodes are used to provide computing, storage, and network resources, and can be, but are not limited to, servers, edge computing devices, etc.

[0027] Optionally, the computing power service request is used to request the computing power resources required for packet processing. The computing power service request may carry the specific requirements of the user terminal for computing power resources or services, and may also include the computing power service preferences of the user terminal.

[0028] Step S104, based on the computing power service request, determining the target computing power service node with the highest computing power factor from the target computing power routing table, and the routing path from the user terminal to the target computing power service node; wherein, the target computing power routing table stores the corresponding relationship between the computing power factor and the computing power service node, and the routing path to the corresponding computing power service node; the computing power factor is used to measure the computing power service quality of the corresponding computing power service node.

[0029] Optionally, in step S104, after receiving the computing power service request from the user terminal, the computing power gateway will, based on the information in the request, search for the computing power service node with the highest computing power factor in its target computing power routing table. The target computing power routing table is a key data structure that stores the correspondence between the computing power factor and the computing power service node, as well as the routing path information from the user terminal to these service nodes. This routing path is the optimal routing path from the user terminal to the target computing power service node. The computing power factor is used to measure the quality of the service provided by the computing power service node. The computing power gateway will compare the computing power service quality of different computing power service nodes according to the computing power factors stored in the target computing power routing table. A node with a higher computing power factor means better service quality. For a specific computing power service request, such a node can provide better computing resources, lower network latency, or a service configuration that better meets the user's preferences. Once the target computing power service node with the highest computing power factor is determined, the computing power gateway will then search for the routing path from the user terminal to this computing power service node to ensure that the user request can be forwarded to the target node through the optimal path, thereby efficiently and with low latency processing the user task. The above process makes full use of the capabilities of the computing power gateway. The announcement of computing power information and the calculation of computing power factors can be achieved by extending the IS-IS protocol, and then a computing power routing table containing computing power factor attributes is generated. The computing power factor is used to search for and schedule computing power service nodes in the target computing power routing table, and to determine the optimal routing path to these computing power service nodes, so as to achieve a rapid response and efficient processing of the user's computing power service request.

[0030] In an alternative embodiment, before determining the target computing power service node with the highest computing power factor in the target computing power routing table based on the computing power service request, and the routing path from the user terminal to the target computing power service node, the method further includes: obtaining the computing power factor corresponding to any one of the multiple computing power service nodes included in the computing power network in the following manner: obtaining the computing power resource ownership of any one computing power service node, and the weight values corresponding to various computing power resources, where the computing power resource ownership includes the quantity of each type of computing power resource owned by any one computing power service node; based on the computing power resource ownership and the weight values corresponding to various computing power resources, obtaining the computing power factor of any one computing power service node; using the method of obtaining the computing power factor of any one computing power service node to obtain the computing power factors corresponding to the multiple computing power service nodes respectively; based on the multiple computing power service nodes, the computing power factors corresponding to the multiple computing power service nodes respectively, and the routing paths corresponding to the multiple computing power service nodes respectively, constructing the target computing power routing table.

[0031] Optionally, in the computing power network, each computing power service node will regularly report its computing power resource ownership to the computing power gateway. At the same time, the computing power gateway is preset with weight values ​​corresponding to a variety of computing power resources, which can reflect the importance of different computing power resources in the comprehensive evaluation of the quality of computing power services. After receiving the computing power resource ownership reported by the computing power service node, the computing power gateway will calculate the computing power factor based on the weight value corresponding to each computing power resource. In the above manner, the computing power gateway calculates the computing power factors of all computing power service nodes in the computing power network. The computing power gateway further constructs a target computing power routing table based on the calculated computing power factor and the routing information of the computing power service node. In the computing power routing table, each computing power service node is assigned a corresponding computing power factor, and the optimal routing path to the computing power service node is also recorded.

[0032] It should be noted that the amount of computing power resources owned is the amount of available resources of each type of computing power resource owned by the computing power service node. The amount of computing power resources owned and the network status are dynamically changing, so the computing power factor and the target computing power routing table will also be updated regularly. The computing power gateway will continue to monitor the resource status updates from the computing power service node. Once a change is detected, it will immediately recalculate the computing power factor and update the computing power information and routing path in the target computing power routing table to reflect the latest network status and computing power service quality.

[0033] Through the above methods, the computing power gateway can perceive the resource capabilities and network status of each computing power service node in the computing power network in real time, and then generate the corresponding target computing power routing table. When the user initiates a computing power service request, the computing power gateway will select the service node with the highest computing power factor from the target computing power routing table, as well as the optimal routing path to the node, so as to efficiently dispatch the user request to the best computing power service node, meeting the user's high-quality demand for computing power services.

[0034] In an optional embodiment, before obtaining the computing power resource ownership of any computing power service node and the weight values ​​corresponding to multiple computing power resources, the method also includes: receiving the computing power resource ownership reported by multiple computing power service nodes respectively.

[0035] Optionally, considering that the amount of computing resources and network conditions are changing dynamically, each of the multiple computing service nodes can report its own computing resources to the computing gateway based on a preset scheduled task. In this way, the computing gateway can perceive the resource capabilities of each computing service node in the computing network in real time and realize dynamic update of the computing routing table.

[0036] Optionally, multiple computing power resources may include, but are not limited to, computing power hardware resources, computing power service resources, and computing power network resources. Among them, the computing power hardware resources are used to indicate the capabilities of the hardware resources of the computing power service nodes, and may include, but are not limited to, computing capabilities, memory capabilities, and storage capabilities. Different capabilities can be characterized by different metrics. The corresponding metrics of the computing power hardware resources are defined in Table 1 below.

[0037] Table 1

[0038]

[0039] The computing power service resources are used to indicate the capabilities of the computing power service nodes to provide services or business support. They can be quantified by metrics such as the computing power service capability (Computing Service Capability), such as the capability grading of the computing power service. The capability grading of the computing power service is used to quantify the capability level of the computing power service nodes to provide services or business support. This capability grading can be based on multiple dimensions, including but not limited to the software and hardware configurations of the computing power nodes, service availability, task processing efficiency, and service scalability, etc. Among them:

[0040] The computing power hardware configuration level quantifies the capability grading of the computing power service according to the hardware configuration of the computing power service nodes. For example, the CPU model, memory size, storage type (solid-state drive or mechanical hard drive), and the type and quantity of GPUs, etc. The computing power service nodes are divided into different hardware configuration levels, such as "low configuration", "standard", "high configuration", and "super" levels. For example, a node with a large number of GPUs and high-speed solid-state drive storage can be classified as the "super" level, while a node with only a basic CPU and a small memory may be classified as the "low configuration" level.

[0041] The service availability level is used to measure the stability and reliability of the computing power service, including the normal running time of the service, the fault recovery speed, and the service interruption frequency, etc. For example, if a computing power service node can provide high availability, such as 99.99% service fault-free time, it is rated as the "high availability" level, while a node with frequent failures or service interruptions is rated as the "low availability" level.

[0042] The task processing efficiency level quantifies the capability grading of the computing power service according to the efficiency of the computing power nodes in executing specific tasks, such as processing speed, concurrent task capability, and the average time to complete tasks, etc. For example, a computing power service node that can quickly process a large amount of data in parallel can be rated as the "high performance" level, while a computing power service node with a slow processing speed or limited concurrent task capability can be rated as the "low performance" level.

[0043] The service scalability level is used to evaluate the expansion ability of computing power service nodes in the face of business growth, including the ability to dynamically adjust resources, the horizontal or vertical scalability of services, and the ability to quickly adapt to new business requirements. For example, a computing power service node that can quickly expand resources to meet business needs can be rated as a "high scalability" level, while a computing power service node with poor scalability can be rated as a "low scalability" level.

[0044] The computing power network resources are used to indicate the network conditions between the links in the computing power network. The metrics can be quantified by one or more of the network delay (Delay time), network bandwidth, packet loss rate, network jitter, network throughput, and link utilization between the computing power gateway and the computing power service node. Among them, network jitter is used to indicate the fluctuation of the arrival time of data packets during network transmission; network throughput represents the amount of data that can be successfully transmitted by the network link per unit time; link utilization represents the ratio of the current traffic of the network link to the maximum transmission capacity.

[0045] In an alternative embodiment, based on the computing power resource ownership and the weight values corresponding to various computing power resources, the computing power factor of any computing power service node is obtained, including: obtaining the total amount of resources corresponding to various computing power resources, where the total amount of resources is used to indicate the total amount of computing power resources of multiple computing power service nodes with the corresponding types; determining the resource proportion corresponding to various computing power resources based on the computing power resource ownership and the total amount of resources corresponding to various computing power resources; and obtaining the computing power factor of any computing power service node based on the resource proportion corresponding to various computing power resources and the weight values corresponding to various computing power resources.

[0046] Optionally, the computing power gateway first obtains the total amount of resources for each type of computing power resource (such as CPU, memory, storage, computing power service ability, and network latency) in the entire computing power network. The total amount of resources refers to the sum of a certain type of computing power resource on all computing power service nodes. For example, the total number of CPUs or the total memory capacity of all computing power service nodes. Obtaining the total amount of resources helps the computing power gateway comprehensively understand the distribution of computing power resources in the network and provides a global perspective for subsequent calculation of computing power factors. After obtaining the total amount of resources, the computing power gateway calculates the proportion of the computing power resource ownership of each computing power service node in the total resource amount, that is, the resource ratio. The resource ratio can reflect the share of a specific computing power service node in the overall computing power network for different computing power resources. For example, the proportion of the number of CPUs of a certain computing power service node in the total number of CPUs of all nodes. Further, based on the resource ratio and a preset weight value, the computing power factor of each computing power service node is calculated. The weight value is used to represent the relative importance of different computing power resources in the final evaluation of computing power service quality, and users can flexibly configure it according to business requirements. Multiply the resource ratio (i.e., the ratio of computing power resource ownership to the total amount of resources) by the corresponding weight value, and then sum up the weighted results of all computing power resources to obtain the final computing power factor. It should be noted that for network latency, its weight value is negative, indicating the characteristic that the lower the latency, the higher the computing power factor. The computing power factor can comprehensively reflect the computing power service quality of the computing power service node, including aspects such as the sufficiency of hardware resources, service ability, and network conditions. Using the computing power factor, the computing power gateway can quickly identify the computing power service node with the best computing power service quality in the network and provide the best computing power service experience for users. Once the computing power factors of all computing power service nodes are calculated, the computing power gateway constructs a target computing power routing table based on these computing power factors and the routing path information from the user side to the computing power service node. This routing table not only includes the correspondence between the computing power factor and the service node but also records the optimal path to the service node, providing a basis for the intelligent scheduling of user requests.

[0047] Through the above method, the computing power factor of the computing power service node can be accurately calculated, and then the target computing power routing table can be constructed to achieve intelligent routing selection and path optimization based on the computing power service quality. This mechanism can ensure that the computing power gateway can provide the computing power service that best meets the needs of users according to the user's computing power service preferences and the real-time state of the network, and at the same time transmit data through the optimal path to improve the efficiency of the computing power network and the user experience.

[0048] Optionally, the computing power resource ownership is the available resource amount of each computing power resource among the multiple computing power resources owned by the computing power service node. According to the weight of each computing power resource information configured on the computing power gateway, weighted calculation is performed: resource weight value * (available resource amount / total resource). Examples of the weight configuration of each computing power resource information are shown in Table 2 below.

[0049] Table 2

[0050]

[0051]

[0052] The weight value of network latency is negative, and the weight values corresponding to other computing power resource metrics are positive. According to the above example, the calculation method of the computing power factor is as follows: 0.2 * (3 / 8) + 0.2 * (480 / 1024) + 0.1 * (4096 / 16384) + 0.4 * (5 / 30) - 0.1 * (320 / 2000).

[0053] In an alternative embodiment, obtaining the computing power resource ownership of any computing power service node includes: receiving a Link State Protocol Data Unit (LSP) packet sent by a first computing power gateway based on the Intermediate System to Intermediate System (IS-IS) protocol, where the LSP packet carries various computing power resources, and the first computing power gateway is a computing power gateway close to the side of the computing power service node; parsing and processing the LSP packet to obtain the computing power resource ownership of any computing power service node.

[0054] Optionally, the computing power gateway (especially the computing power gateway on the user side) receives LSP packets sent by the first computing power gateway (the computing power gateway closer to the computing power service node side) based on the IS-IS protocol. These LSP packets contain the computing power resource information of the computing power service node, such as the number of CPUs, memory capacity, storage capacity, computing power service capacity grading, and network latency. The computing power resource information is encapsulated in a custom Computing-aware TLV (Type-Length-Value) for transmission through the standard IS-IS protocol. After receiving the LSP packets, the user-side computing power gateway parses and processes the packets to extract the computing power resource ownership information of the computing power service node. The parsing process involves identifying the Computing-aware TLV in the packets and reading the computing power resource data carried therein, including but not limited to parameters such as the number of CPUs, memory capacity, storage space, and computing power service capacity. Through the computing power resource information obtained by parsing the LSP packets, the user-side computing power gateway can update its internal computing power resource database in real time to ensure that the resource status of the computing power service node it holds is the latest. After successfully decoding and updating the computing power resource information, the user-side computing power gateway calculates the computing power factor of each computing power service node based on the latest resource status and a preset weight value. The weight value can reflect the importance of different computing power resources, and through weighted calculation, the overall service quality of the computing power service node can be comprehensively evaluated. The calculation result of the computing power factor is used to construct and maintain the target computing power routing table. This routing table not only records the correspondence between the computing power factor and the computing power service node but also contains the optimal routing paths to these service nodes. With the continuous update of the computing power resource information, the computing power factor and the target computing power routing table will also be dynamically adjusted to reflect the latest and most accurate resource status and path information of the computing power service nodes in the computing power network.

[0055] Optionally, the routing calculation ability of the IS-IS protocol can be utilized to ensure that the service node with the highest computing power factor can be accessed through the most optimized path. The corresponding generated target computing power routing table is a computing power routing table constructed based on the IS-IS protocol. Still as Figure 2 shown, before executing step S104, after the computing power service scheduling platform sends the computing power resource information (including the quantity of each type of computing power resource among multiple computing power resources) to the computing power gateway, the computing power resource information announcement is completed with the help of the IS-IS protocol. The computing power gateway calculates the computing power resource information into a computing power factor according to the set computing power resource weight to form a computing power routing table, thereby completing the full-network computing power perception.

[0056] Optionally, to support the IS-IS protocol for extended computing power routing, the Type-Length-Value (TLV) triple is a Computing-aware TLV. Here, the TLV is used to carry specific types of information in the Link State Protocol Data Unit (LSP) packet. To support the extended computing power routing function, the Computing-aware TLV is designed to carry computing power information related to computing resources, computing power services, and network conditions, enabling the IS-IS protocol to sense and advertise this information, thus supporting routing selection based on computing power factors. By parsing this TLV, the computing power gateway can understand the resource status and network conditions of computing power service nodes in the network, and then generate a computing power routing table to achieve dynamic awareness and intelligent scheduling of computing power resources. Figure 3 is an optional schematic diagram of the IS-IS protocol structure according to an embodiment of the present invention, where:

[0057] Type represents the TLV type, which is used to define the computing power routing TLV as the Computing-aware computing power routing type, and the information carried is computing power routing information.

[0058] Length represents the TLV length, which is used to define the total length of the computing power routing type TLV.

[0059] rsv represents the reserved field, which is used to reserve space for future protocol extensions or specific purposes.

[0060] Flags represents the flag bit, which is used to extend some features in the future.

[0061] Compute node IP is the IP address of the computing power service node, which represents the IP address of the computing power resource that can provide computing power services for users in the computing power service node, with a length of 32 bytes (bits).

[0062] Cpu Number represents the number of CPUs of the computing power service node, which is used to measure the computing power of the computing power node, with a length of 32 bits.

[0063] Memory size represents the memory size of the computing power service node, with a length of 32 bits.

[0064] Storage represents the external storage capacity of the computing power service node, with a length of 32 bits.

[0065] Computing Service Capability represents the service capability level of the computing power service node, with a length of 32 bits.

[0066] Delay time represents the time delay from the computing power service node to the computing power gateway, with a length of 32 bits.

[0067] In an alternative embodiment, before determining the target computing power service node with the highest computing power factor from the target computing power routing table and the routing path from the user side to the target computing power service node based on the computing power service request, the method further includes: determining the target computing power service preference of the user side based on the computing power service request, where the target computing power service preference is used to indicate the preference degree of the user side for different types of computing power resources; determining the target computing power routing table that matches the target computing power service preference from multiple groups of computing power routing tables, where the multiple groups of computing power routing tables correspond to different computing power service preferences.

[0068] Optionally, when the user initiates a computing power service request, the computing power gateway first needs to determine the target computing power service preference carried or implied by the request. The preference can reflect the user's emphasis on different computing power resources (such as CPU, memory, storage, service capabilities, and network latency), and can be determined through user configuration or automatic learning. The user can specify a specific computing power service preference by configuring an identifier (Compute Service ID, CSID). CSID is an identifier that is associated with a set of weight values, and the weight values are used to define the user's preference degree for different computing power resources. For example, CSID1 may place more emphasis on the number of CPUs, while CSID2 may place more emphasis on memory capabilities. The computing power gateway maintains multiple groups of computing power routing tables, and each group of routing tables corresponds to a different computing power service preference configuration. After determining the user's target computing power service preference, find the group of computing power routing tables that best matches this preference configuration from the multiple groups of computing power routing tables maintained. When each group of computing power routing tables is generated, the computing power factor is calculated based on a specific computing power service preference configuration. Therefore, different computing power service preferences will result in different computing power factors and optimal path selections. Once the target computing power routing table that matches the user request is determined, the computing power gateway can look up the service node with the highest computing power factor in this table and the optimal routing path from the user side to this service node. The computing power factor can comprehensively evaluate the resource status and service quality of the computing power service node. A high-factor node means that under the current preference configuration, it provides better service capabilities. For example, if the user prefers CSID1, the computing power gateway will look up the computing power routing table corresponding to CSID1 and select the computing power service node with the highest computing power factor from it. This indicates that in terms of the computing power resources that the user values (such as the number of CPUs), this computing power service node can provide the best service.

[0069] In the above manner, the computing power gateway can flexibly respond to different users' different demands for computing power resources, ensuring that users can be scheduled to the computing power nodes that provide the best quality services according to their service preferences, and at the same time select the optimal path for data transmission. This can not only improve the service experience of users, but also optimize the resource utilization rate and overall performance of the computing power network.

[0070] In an optional embodiment, before determining the target computing power routing table that matches the target computing power service preference from multiple groups of computing power routing tables, the method further includes: obtaining multiple computing power service preferences, and weight ratio combinations respectively corresponding to the multiple computing power service preferences, where the weight ratio combination is used to indicate the weight values respectively allocated to multiple computing power resources under the corresponding computing power service preference; constructing multiple groups of computing power routing tables based on the weight ratio combinations respectively corresponding to the multiple computing power service preferences.

[0071] Optionally, construct multiple groups of computing power routing tables based on the weight ratio combinations respectively corresponding to multiple computing power service preferences and the computing power resource ownerships respectively corresponding to multiple computing power service nodes. Specifically, the computing power gateway first needs to obtain multiple pre-defined computing power service preference configurations. The preference configuration is a set of rules pre-defined by users or system administrators, which describes the preference weights in the selection of computing power resources, such as CSID1, CSID2, and CSID3, etc. Each CSID corresponds to a different set of weight ratios, which can reflect different preferences for computing power resources in different business scenarios. Each computing power service preference (such as CSID1, CSID2, CSID3) corresponds to a specific weight ratio combination, which defines which computing power resources (such as the number of CPUs, memory capacity, storage capacity, etc.) are given higher priorities or weights when calculating computing power factors. For example, CSID1 may allocate more weights to the number of CPUs, while CSID2 may consider memory capacity more. The computing power gateway constructs a group of computing power routing tables for each computing power service preference according to the multiple computing power service preferences obtained and their corresponding weight ratio combinations. This process involves the calculation of computing power factors, that is, weighted calculation is performed according to the weight of each computing power information configured on the computing power gateway, and finally the computing power factor is obtained. Each group of computing power routing tables contains the computing power factors calculated based on different preferences and the corresponding routing information, so as to support the optimal selection of computing power resources and service nodes under different preferences. When a user request arrives, the computing power gateway determines the target computing power service preference (such as CSID1) according to the user request, and then selects a group of computing power routing tables that match this preference from the multiple groups of computing power routing tables constructed before. In this group of computing power routing tables, the computing power gateway further determines the optimal computing power service node and the routing path from the user side to this node according to the level of the computing power factor.

[0072] In the above manner, multiple groups of computing power routing tables are pre-constructed and maintained, and each group of routing tables corresponds to different combinations of computing power service preferences and weight ratios. This enables quick and intelligent matching of the user's computing power service preferences when facing user requests, and optimally selects the most suitable computing power service nodes and routing paths for the user, thereby meeting the personalized needs of different users and improving the scheduling efficiency of computing power resources and the user's service experience.

[0073] Optionally, in the routing calculation process of the IS-IS routing protocol, the calculated computing power factors are carried in the computing power route, and a set of computing power routing tables including the attributes of the computing power factors is generated. Figure 4 FIG. is a schematic flow chart of generating a computing power routing table based on the IS-IS protocol according to an optional embodiment of the present invention. As Figure 4 shown, the specific process includes: parsing the computing power resource information carried in the LSP packet, that is, the computing power resource ownership of the computing power service node; generating computing power factors based on the weight combination corresponding to the computing power service preference and the computing power resource information in the LSP packet; based on the identification IP of the computing power service node, performing routing calculation using the Shortest Path First (SPF) algorithm of the IS-IS routing protocol, and using the computing power factor as the attribute of the routing prefix to generate a computing power routing table with the attributes of the computing power factor.

[0074] In an optional embodiment, the multiple computing power resources at least include: computing power hardware resources, computing power service resources, and computing power network resources, and the multiple computing power service preferences at least include preferences for computing power hardware resources, preferences for computing power service resources, and preferences for computing power network resources.

[0075] Optionally, the computing power service nodes in the computing power network can provide computing power hardware resources or computing power service resources. Different computing power service nodes provide different capabilities, and the perceived and advertised computing power resources are also different. In this embodiment, the computing power resource information for IS-IS protocol advertisement is defined from three aspects: computing power hardware resources, computing power service resources, and computing power network resources. These three types of computing power resource information are calculated as computing power factors according to the weights, and a computing power routing table with the attributes of the computing power factors is generated. If multiple computing power service preferences are configured at the computing power gateway, multiple computing power factors and multiple groups of computing power routing tables are generated based on different weight ratios.

[0076] Optionally, the preference for computing power hardware resources can be understood as being concerned about the number of CPUs, the preference for computing power service resources can be understood as being concerned about the memory capacity, and the preference for computing power network resources can be understood as being concerned about the network latency. For example, CSID is used to identify a group of computing power service preferences, and the specific format is shown in Table 3. Three examples of computing power service preferences are illustrated as follows:

[0077] Table 3

[0078]

[0079] It can be seen that CSID1 is concerned about the number of CPUs, CSID2 is concerned about the memory capacity, and CSID3 is concerned about the network latency. Based on these three computing power service preferences, three computing power routing tables can be generated for the computing power resources from the same computing power service node.

[0080] Figure 5 It is a schematic flowchart of a process for generating a computing power routing table corresponding to a computing power service node under multiple computing power service preferences according to an embodiment of the present invention. As Figure 5 shown, the computing power resource information corresponding to the same computing power service node is scored based on three different computing power service preferences (i.e., preference one, preference two, and preference three). Based on the weight values of the three different computing power service preferences and the computing power resource information corresponding to the computing power service node, weighted calculations are performed to obtain multiple computing power factors obtained under different computing power service preferences, which are respectively used for constructing computing power routing tables under different computing power service preferences.

[0081] Figure 6 It is a schematic flowchart of a process for generating a computing power routing table corresponding to a computing power service node under multiple computing power service preferences according to an embodiment of the present invention. As Figure 6 shown, the computing power resource information corresponding to each computing power service node among multiple computing power service nodes is scored based on three different computing power service preferences (i.e., preference one, preference two, and preference three). Based on the weight values of the three different computing power service preferences and the computing power resource information corresponding to each computing power service node, weighted calculations are performed to obtain multiple computing power factors obtained by each computing power service node under different computing power service preferences, which are respectively used for constructing computing power routing tables under different computing power service preferences. In this case, "computing power resource information one" needs to be configured based on three computing power service preferences to calculate three computing power routing tables (i.e., computing power route one, computing power route two, and computing power route three). The same operation is performed on the computing power resource information two to calculate the corresponding three computing power routing tables (i.e., computing power route four, computing power route five, and computing power route six). At this time, the number of computing power routing table entries on the computing power gateway is: the number of computing power resource information * the number of computing power service preferences. Each computing power routing table includes the computing power factors corresponding to multiple computing power service nodes respectively under the corresponding computing power service preference, and the routing path to the computing power service node.

[0082] Figure 7 It is a schematic flowchart of an optional computing power routing process according to an embodiment of the present invention. As Figure 7As shown in the figure, upon receiving the computing power service request sent by the user terminal to request the distribution of the data plane hardware table, if the computing power service preference of the user indicated in the computing power service request is to be concerned about the number of CPUs, it is determined to use the computing power routing table constructed based on CSID1 to determine the corresponding computing power reason path for the user terminal; the protocol module indexes to the corresponding entry according to CSID1, selects a computing power route, and learns from querying the computing power routing table constructed based on CSID1 that the computing power factor corresponding to "Computing Power Route 1" is the largest, so "Computing Power Route 1" is used as the selected computing power route; the data plane is distributed according to the selected computing power route, and thereafter the user terminal directly addresses and schedules to the specified computing power service node through the hardware routing table.

[0083] Through the above steps S102 to S104, the correspondence between the computing power factor and the computing power service node in the target computing power routing table can be constructed and utilized, and the purpose of automatically selecting the node with the highest computing power service quality and its routing path according to the user request can be achieved, so as to achieve the technical effect of improving the efficiency and accuracy of the selection of the computing power service node and the path optimization, and further solve the technical problems of inaccurate and inefficient selection of the computing power service node and path optimization in the related art.

[0084] Based on the above embodiments and alternative embodiments, the present invention proposes an alternative implementation manner of the computing power routing method, and this method is applied to a Figure 2 computing power network networking based on the IS-IS protocol as shown in the figure. The computing power gateway is an important network element device in the computing power network. Based on the three core capabilities of computing network perception, computing power routing, and computing network transmission guarantee, the computing power gateway realizes the optimal scheduling and transmission of computing network traffic, and is an important network infrastructure for realizing the integrated scheduling of the computing power network. The overall working process of the computing power network service is as follows:

[0085] S21, the user terminal registers a service request with the computing power service scheduling platform;

[0086] S22, the computing power service scheduling platform selects a computing power resource pool for service deployment;

[0087] S23, the computing power service scheduling platform sends the computing power resource information to the computing power gateway;

[0088] S24, the computing power gateway generates a computing power routing table through computing power information advertisement to complete the computing power perception of the whole network.

[0089] S25, the user terminal requests service traffic to enter the computing power gateway.

[0090] S26, the computing power gateway schedules the user traffic to the optimal and best computing power service node for processing according to the user service requirements (i.e., the computing power service request).

[0091] Figure 8It is a schematic diagram of an optional IS-IS protocol message interaction process according to an embodiment of the present invention. Through the newly added computing power routing TLV in the LSP message defined in the foregoing embodiment, announcements and synchronizations are performed. The specific implementation process is as follows:

[0092] S71, An IS-IS neighbor relationship is established between the computing power gateway on the user side (i.e., computing power gateway A) and the computing power gateway on the computing power resource side (i.e., computing power gateway B);

[0093] S72, The computing power gateway A and the computing power gateway B send database description messages (CSNP messages) to each other, send the summary of their own computing power routing information, and synchronize information with the peer;

[0094] S73, The computing power gateway A sends a request message (PSNP message) to request the complete information of the computing power routing LSP that does not exist in its own database;

[0095] S74, The peer computing power gateway B replies to the computing power gateway A with an update message, including the complete information of the computing power routing LSP;

[0096] S75, After receiving the computing power routing LSP, the computing power gateway A sends a LSP confirmation message (PSNP) for confirmation, stating that it has received the LSP message;

[0097] S76, When the computing power information of the computing power gateway B changes, a computing power routing update is performed, and an LSP update message is sent to send the updated computing power routing LSP to the computing power gateway B;

[0098] S77, After receiving the update message, the computing power gateway B sends a PSNP confirmation message to the computing power gateway A for confirmation.

[0099] Figure 9 It is a schematic diagram of an optional IS-IS computing power routing table generation process according to an embodiment of the present invention. As Figure 9 shown, it specifically includes:

[0100] S81, The computing power service node reports the computing power information to the computing power gateway B through methods such as computing power reporting or computing power discovery;

[0101] S82, The computing power gateway B generates an LSP message containing computing power resource information;

[0102] S83, The computing power gateway B floods and sends the LSP message to the corresponding computing power gateway A on the user side via the computing power routing node;

[0103] S84, The computing power gateway A parses the received computing power information LSP message, obtains the computing power factor, and generates a computing power routing table through the IS-IS protocol. Among them, the acquisition method of the computing power routing table is the same as the foregoing, and will not be elaborated here;

[0104] S85, when the user accesses the computing power service, a business request (i.e., computing power service request) is sent;

[0105] S86, computing power gateway A matches the computing power routing table according to the business request. When there are multiple computing power service nodes, the corresponding computing power service node can be selected by the computing power factor;

[0106] S87, forward the service request to the corresponding computing power gateway node B. The computing power service node receives the service request and forwards the service traffic to the corresponding computing power gateway A, which then sends it to the user end.

[0107] Figure 10 FIG. 1 is a schematic diagram of an optional computing service scheduling process for different user terminals based on different business requirements according to an embodiment of the present invention. Figure 10 As shown in the figure, when a user initiates a request for computing service CSID1 1.1.1.1, after reaching the computing gateway, the first packet is sent to the CPU, triggering the protocol routing; the computing gateway finds that there are two computing routes in the computing routing table under the current CSID1 index. By comparing the results of the computing factor, computing route 1 is selected, and this computing route is directly sent to the data plane; after that, when the user initiates a request for computing service CSID1, after reaching the computing gateway, it is directly forwarded through the hardware of the data plane and reaches the final computing service node 1.

[0108] In summary, in this embodiment, a custom Computing-aware TLV is added, and the IS-IS protocol is used to transmit computing power information; the ISIS protocol is added to calculate the computing power routing process and generate a computing power routing table based on the computing power factor; a method for calculating the computing power information into the computing power factor is added; a computing power service preference configuration is added, and different computing power routing tables are generated for the same set of computing power information based on different weight ratios to match the business needs of different users.

[0109] Specifically, after receiving the computing power resource information, the computing power gateway needs to notify the computing power information through the routing protocol (in this embodiment, the IS-IS protocol is used to expand the computing power resource information). After receiving the computing power information, each computing power gateway parses the computing power information and generates a computing power routing table in combination with the computing power information weight of the computing power gateway (based on the weights of each computing power information configuration, it can be flexibly configured and changed at each gateway node, and the calculated value is hereinafter referred to as the computing power factor). The computing power factor is an attribute of the computing power routing table, and the computing power routing table is used to select the optimal path to reach the designated computing power service node; at the same time, when multiple computing power service preferences are configured under the computing power gateway, multiple computing power routing tables are generated for the same computing power service node. When planning services, the computing power factor can be used as a factor in user route selection, so that different users can choose computing power routing tables based on different business needs, and the business can be dispatched to the service node along the designated path.

[0110] In this embodiment, a computing power routing device is further provided. This device is used to implement the above-mentioned embodiments and preferred implementation manners, and those that have been described will not be repeated. As used hereinafter, the terms "module" and "device" can be a combination of software and / or hardware that can achieve a predetermined function. Although the devices described in the following embodiments are preferably implemented in software, implementation in hardware, or a combination of software and hardware is also possible and contemplated.

[0111] According to an embodiment of the present invention, an apparatus embodiment for implementing the above-mentioned computing power routing method is also provided. Figure 11 It is a schematic structural diagram of a computing power routing device according to an embodiment of the present invention. As Figure 11 shown, the above-mentioned computing power routing device includes: a request receiving module 1100 and a routing path determining module 1102, where:

[0112] The request receiving module 1100 is configured to receive a computing power service request sent by the user terminal;

[0113] The routing path determining module 1102 is connected to the request receiving module 1100 and is configured to determine, based on the computing power service request, a target computing power service node with the highest computing power factor from the target computing power routing table, and a routing path from the user terminal to the target computing power service node; wherein, the target computing power routing table stores the corresponding relationship between the computing power factor and the computing power service node, and the routing path to the corresponding computing power service node; the computing power factor is used to measure the computing power service quality of the corresponding computing power service node.

[0114] In the embodiment of the present invention, by setting the request receiving module 1100 to receive the computing power service request sent by the user terminal; the routing path determining module 1102 is connected to the request receiving module 1100 and is configured to determine, based on the computing power service request, a target computing power service node with the highest computing power factor from the target computing power routing table, and a routing path from the user terminal to the target computing power service node; wherein, the target computing power routing table stores the corresponding relationship between the computing power factor and the computing power service node, and the routing path to the corresponding computing power service node; the computing power factor is used to measure the computing power service quality of the corresponding computing power service node, the purpose of constructing and utilizing the corresponding relationship between the computing power factor and the computing power service node in the target computing power routing table and automatically selecting the node with the highest computing power service quality and its routing path according to the user request is achieved, thereby achieving the technical effect of improving the efficiency and accuracy of computing power service node selection and path optimization, and further solving the technical problems of inaccurate and inefficient computing power service node selection and path optimization in the related art.

[0115] It should be noted that the above-mentioned modules can be implemented by software or hardware. For example, for the latter, it can be implemented in the following ways: the above-mentioned modules can be located in the same processor; or, the above-mentioned modules are located in different processors in any combination.

[0116] It should be noted here that the above-mentioned request receiving module 1100 and routing path determining module 1102 correspond to steps S102 to S104 in the embodiment. The examples and application scenarios implemented by the above-mentioned modules and the corresponding steps are the same, but are not limited to the content disclosed in the above-mentioned embodiment. It should be noted that the above-mentioned modules can run in a computer terminal as part of the device.

[0117] It should be noted that the optional or preferred implementation manners of this embodiment can be referred to the relevant descriptions in the embodiment, and will not be repeated here.

[0118] The above-mentioned computing power routing device may further include a processor and a memory. The above-mentioned request receiving module 1100, routing path determining module 1102, etc. are all stored in the memory as program modules, and the corresponding functions are implemented by the processor executing the above-mentioned program modules stored in the memory.

[0119] The processor contains a kernel, and the kernel retrieves the corresponding program module from the memory. One or more kernels can be set. The memory may include non-permanent memory in a computer-readable medium, random access memory (RAM) and / or non-volatile memory in the form of, for example, read-only memory (ROM) or flash memory (flash RAM). The memory includes at least one memory chip.

[0120] According to an embodiment of the present application, an embodiment of a non-volatile storage medium is also provided. Optionally, in this embodiment, the above-mentioned non-volatile storage medium includes a stored program, wherein when the above-mentioned program runs, it controls the device where the above-mentioned non-volatile storage medium is located to execute any one of the above-mentioned computing power routing methods.

[0121] Optionally, in this embodiment, the above-mentioned non-volatile storage medium can be located in any one of the computer terminals in a computer terminal group in a computer network, or in any one of the mobile terminals in a mobile terminal group. The above-mentioned non-volatile storage medium includes a stored program.

[0122] Optionally, when the program is running, control the device where the non-volatile storage medium is located to perform the following functions: receive a computing power service request sent by the client; based on the computing power service request, determine the target computing power service node with the highest computing power factor from the target computing power routing table, and the routing path from the client to the target computing power service node; wherein, the target computing power routing table stores the correspondence between the computing power factor and the computing power service node, and the routing path to the corresponding computing power service node; the computing power factor is used to measure the computing power service quality of the corresponding computing power service node.

[0123] According to an embodiment of the present application, an embodiment of a processor is further provided. Optionally, in this embodiment, the above-mentioned processor is used to run a program, wherein, when the above-mentioned program runs, it executes any one of the above-mentioned computing power routing methods.

[0124] According to an embodiment of the present application, an embodiment of a computer program product is further provided. When executed on a data processing device, it is adapted to execute a program initialized with the steps of any one of the above-mentioned computing power routing methods.

[0125] Optionally, when the above-mentioned computer program product is executed on a data processing device, it is adapted to execute a program initialized with the following method steps: receive a computing power service request sent by the client; based on the computing power service request, determine the target computing power service node with the highest computing power factor from the target computing power routing table, and the routing path from the client to the target computing power service node; wherein, the target computing power routing table stores the correspondence between the computing power factor and the computing power service node, and the routing path to the corresponding computing power service node; the computing power factor is used to measure the computing power service quality of the corresponding computing power service node.

[0126] An embodiment of the present invention provides an electronic device, which includes a processor, a memory, and a program stored on the memory and executable on the processor. When the processor executes the program, the following steps are implemented: receive a computing power service request sent by the client; based on the computing power service request, determine the target computing power service node with the highest computing power factor from the target computing power routing table, and the routing path from the client to the target computing power service node; wherein, the target computing power routing table stores the correspondence between the computing power factor and the computing power service node, and the routing path to the corresponding computing power service node; the computing power factor is used to measure the computing power service quality of the corresponding computing power service node.

[0127] The above order of the embodiments of the present invention is only for description and does not represent the advantages or disadvantages of the embodiments.

[0128] In the above embodiments of the present invention, the descriptions of each embodiment have their own emphases. For the parts not detailed in a certain embodiment, reference can be made to the relevant descriptions of other embodiments.

[0129] In several embodiments provided by the present application, it should be understood that the disclosed technical content can be implemented in other ways. Among them, the device embodiments described above are merely illustrative. For example, the above-mentioned module division can be a logical function division. In actual implementation, there can be other division methods. For example, multiple modules or components can be combined or integrated into another system, or some features can be ignored or not executed. Another point is that the displayed or discussed coupling, direct coupling, or communication connection between each other can be through some interfaces. The indirect coupling or communication connection of modules or modules can be in an electrical or other form.

[0130] The modules described above as separate components may or may not be physically separated. The components displayed as modules may or may not be physical modules, that is, they can be located in one place, or they can be distributed to multiple modules. Some or all of the modules can be selected according to actual needs to achieve the purpose of the solution of this embodiment.

[0131] In addition, in each embodiment of the present invention, each functional module can be integrated in a processing module, or each module can exist physically alone, or two or more modules can be integrated in one module. The above-mentioned integrated modules can be implemented in the form of hardware or in the form of software functional modules.

[0132] If the above-mentioned integrated module is implemented in the form of a software functional module and sold or used as an independent product, it can be stored in a computer-readable non-volatile storage medium. Based on such an understanding, the technical solution of the present invention, in essence, or the part that contributes to the prior art, or all or part of this technical solution, can be embodied in the form of a software product. This computer software product is stored in a non-volatile storage medium and includes several instructions for causing a computer device (which can be a personal computer, a server, or a network device, etc.) to execute all or part of the steps of the methods in each embodiment of the present invention. And the aforementioned non-volatile storage medium includes: USB flash drives, read-only memories (ROMs), random access memories (RAMs), mobile hard disks, magnetic disks, or optical discs, etc., which can store program codes.

[0133] The above is only the preferred embodiment of the present invention. It should be noted that for those of ordinary skill in the art, without departing from the principle of the present invention, several improvements and refinements can be made, and these improvements and refinements should also be regarded as the protection scope of the present invention.

Claims

1. A computing power routing method, characterized in that: include: Receive the computing power service request sent by the user; Based on the computing power service request, determine the target computing power service node with the highest computing power factor from the target computing power routing table, and the routing path from the user terminal to the target computing power service node; Among them, the target computing power routing table stores the correspondence between the computing power factor and the computing power service node, as well as the routing path leading to the corresponding computing power service node; the computing power factor is used to measure the computing power service quality of the corresponding computing power service node.

2. The method according to claim 1, characterized in that: Before determining the target computing power service node with the highest computing power factor and the routing path from the user terminal to the target computing power service node from the target computing power routing table based on the computing power service request, the method further includes: The computing power factor corresponding to any computing power service node among the multiple computing power service nodes included in the computing power network is obtained in the following way: Obtain the amount of computing resources owned by any computing service node, and the weight values ​​corresponding to the multiple computing resources, wherein the amount of computing resources owned includes the capacity of each of the multiple computing resources owned by any computing service node; Based on the amount of computing resources owned and the weight values ​​corresponding to the multiple computing resources, obtain the computing power factor of any computing power service node; Obtain the computing power factors corresponding to the multiple computing power service nodes respectively by adopting the method of obtaining the computing power factor of any computing power service node; The target computing power routing table is constructed based on the multiple computing power service nodes, the computing power factors respectively corresponding to the multiple computing power service nodes, and the routing paths respectively corresponding to the multiple computing power service nodes.

3. The method according to claim 2, characterized in that The obtaining the amount of computing resources owned by any computing service node includes: Receiving a link state protocol data unit LSP message sent by the first computing power gateway based on the intermediate system to intermediate system IS-IS protocol, wherein the LSP message carries the multiple computing power resources, and the first computing power gateway is a computing power gateway close to the computing power service node side; The LSP message is parsed and processed to obtain the computing power resource ownership of any computing power service node.

4. The method according to claim 2, characterized in that: The obtaining of the computing power factor of any computing power service node based on the amount of computing power resources owned and the weight values ​​corresponding to the multiple computing power resources respectively includes: Obtaining the total amount of resources corresponding to the multiple computing resources respectively, wherein the total amount of resources is used to indicate the total amount of computing resources of corresponding types owned by the multiple computing service nodes; Based on the amount of computing resources owned and the total amount of resources corresponding to the multiple computing resources, determine the resource proportions corresponding to the multiple computing resources respectively; Based on the resource proportions corresponding to the multiple computing resources and the weight values ​​corresponding to the multiple computing resources, the computing power factor of any computing power service node is obtained.

5. The method according to claim 1, characterized in that Before determining the target computing power service node with the highest computing power factor and the routing path from the user terminal to the target computing power service node from the target computing power routing table based on the computing power service request, the method further includes: Based on the computing power service request, determining a target computing power service preference of the user terminal, wherein the target computing power service preference is used to indicate the degree of preference of the user terminal for different types of computing power resources; Determine the target computing power routing table that matches the target computing power service preference from multiple groups of computing power routing tables, wherein the multiple groups of computing power routing tables correspond to different computing power service preferences.

6. The method according to claim 5, characterized in that Before determining the target computing power routing table that matches the target computing power service preference from the multiple groups of computing power routing tables, the method further includes: Obtain multiple computing power service preferences and weight ratio combinations corresponding to the multiple computing power service preferences, wherein the weight ratio combination is used to indicate the weight values ​​of the multiple computing power resources respectively allocated under the corresponding computing power service preferences; Based on the weight ratio combinations corresponding to the multiple computing power service preferences, the multiple groups of computing power routing tables are constructed.

7. The method according to claim 6, characterized in that The multiple computing power resources include at least: computing power hardware resources, computing power service resources, and computing power network resources. The multiple computing power service preferences include at least preferences for the computing power hardware resources, preferences for the computing power service resources, and preferences for the computing power network resources.

8. The method according to claim 2, characterized in that: Before obtaining the computing power resource ownership of any computing power service node and the weight values ​​corresponding to the plurality of computing power resources, the method further includes: Receive the computing power resource ownership reported by the multiple computing power service nodes respectively.

9. An electronic device, characterized in that: It includes one or more processors and a memory, wherein the memory is used to store one or more programs, wherein when the one or more programs are executed by the one or more processors, the one or more processors implement the computing power routing method described in any one of claims 1 to 8.

10. A computer program product, comprising a computer program, characterized in that When the computer program is executed by a processor, the steps of the computing power routing method described in any one of claims 1 to 8 are implemented.

Images