Business processing methods, devices, electronic equipment and computer program products

Abstract

This application provides a service processing method, apparatus, electronic device, and computer program product. The method includes: determining a target computing node for processing a target customer's service; selecting a preset number of baseline routing nodes from multiple routing nodes, using the connectivity cost relative to the target computing node as a dimension; using each baseline routing node as an initial cluster center; and clustering all non-baseline routing nodes from the multiple routing nodes, using the target attribute of the routing node as a dimension, with different clusters corresponding to different value ranges under the target attribute; determining target clusters whose target attribute value ranges match the target customer's service; and selecting at least two routing nodes from the target clusters, including the corresponding baseline routing node, to configure a transmission path for sending the target customer's service to the target computing node. This application can rationally schedule the computing nodes responsible for processing and the routing nodes responsible for transmission according to the characteristics of the customer service.

Description

Technical Field

[0001] This application relates to the field of cloud computing technology, and in particular to a business processing method, apparatus, electronic device, and computer program product. Background Technology

[0002] Computing power networks are a new type of information infrastructure that allocates and flexibly schedules computing, storage, and network resources on demand among the cloud, network, and edge according to customer business needs, and can provide computing power support for some cloud applications.

[0003] Typically, a computing network comprises multiple compute nodes and multiple routing nodes. Compute nodes handle client requests, while routing nodes form the transmission paths that send client requests to the compute nodes. Elastic scheduling is a crucial metric for evaluating computing network performance. How to efficiently utilize computing network resources and schedule routing and compute nodes for client requests is key to improving user experience. Summary of the Invention

[0004] The purpose of this application is to provide a business processing method, apparatus, electronic device, and computer program product that can reasonably schedule the computing nodes responsible for processing and the routing nodes responsible for transmission according to the characteristics of customer business, and in particular, the routing nodes can send customer business to the computing nodes at a relatively low cost.

[0005] To achieve the above objectives, the embodiments of this application are implemented as follows:

[0006] Firstly, a business processing method is provided, applied to a computing power network, including:

[0007] Identify the target computing nodes in the computing network responsible for processing the target customer's business;

[0008] Using the connectivity cost relative to the target computing node as a dimension, a preset number of reference routing nodes are selected from multiple routing nodes of the computing power network.

[0009] Each of the baseline routing nodes is used as the initial cluster center. All non-baseline routing nodes among the multiple routing nodes are clustered using the target attribute of the routing node as the dimension to obtain the preset number of clusters. Different clusters correspond to different value ranges of the target attribute.

[0010] From the preset number of clusters, determine the target cluster whose value range of the target attribute matches the target customer service, and select at least two routing nodes from the target cluster, including the corresponding baseline routing node, to configure the transmission path for sending the target customer service to the target computing node.

[0011] The target customer service is sent to the target computing node for service processing according to the transmission path.

[0012] Secondly, a service processing device is provided for use in a computing network, comprising:

[0013] The compute node configuration module is used to determine the target compute nodes in the computing network that are responsible for processing the target customer's business.

[0014] The clustering preparation module is used to select a preset number of baseline routing nodes from multiple routing nodes in the computing power network, using the connectivity cost relative to the target computing node as the dimension.

[0015] The clustering execution module is used to take each of the baseline routing nodes as the initial cluster center, and to cluster all non-baseline routing nodes of the plurality of routing nodes with the target attribute of the routing node as the dimension, so as to obtain the preset number of clusters, and different clusters correspond to different value ranges of the target attribute.

[0016] The path determination module is used to determine, from the preset number of clusters, the target cluster whose value range of the target attribute matches the target customer service, and select from the target cluster at least two routing nodes including the corresponding baseline routing node, so as to configure the transmission path for sending the target customer service to the target computing node.

[0017] The service sending module is used to send the target customer service to the target computing node for service processing according to the transmission path.

[0018] Thirdly, embodiments of this application provide an electronic device, including: a processor; and a memory configured to store computer-executable instructions, which, when executed, cause the processor to perform the method described in the first aspect.

[0019] Fourthly, a computer program product is provided, the computer program product including a computer-readable storage medium storing a computer program operable to cause a computer to perform the method described in the first aspect.

[0020] In this embodiment, when a target customer service needs to be processed by the computing power network, the target computing node responsible for processing the target customer service is first determined. Then, a predetermined number of baseline routing nodes are selected from multiple routing nodes in the computing power network, using the connection cost relative to the target computing node as a dimension. Next, each baseline routing node is used as an initial cluster center. Clustering is performed on all non-baseline routing nodes from the multiple routing nodes, using the target attribute of the routing node as a dimension, to obtain clusters with different value ranges of the target attribute. Finally, based on the characteristics of the target customer service, a target cluster with the most matching value range of the target attribute is determined. At least two routing nodes, including the corresponding baseline routing node, are selected from the target cluster to configure the transmission path for sending the target customer service to the target computing node. Since the connection cost of the baseline routing node relative to the target computing node is relatively small, further selecting other non-baseline routing nodes with similar target attributes around the baseline routing node can configure a transmission path with a relatively low overall cost, thereby making more rational use of computing power network resources and allocating the target customer service to computing nodes for processing. Attached Figure Description

[0021] To more clearly illustrate the technical solutions in the embodiments of this application or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments recorded in the embodiments of this application. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0022] Figure 1 This is a schematic diagram of the computing network structure according to an embodiment of this application.

[0023] Figure 2 This is a schematic diagram of a business processing method according to an embodiment of this application.

[0024] Figure 3 This is a schematic diagram of the structure of the service processing device according to an embodiment of this application.

[0025] Figure 4 This is a schematic diagram of the structure of an electronic device according to an embodiment of this application. Detailed Implementation

[0026] Computing power networks are a new type of information infrastructure that allocates and flexibly schedules computing, storage, and network resources on demand among the cloud, network, and edge according to customer business needs, and can provide computing power support for some cloud applications.

[0027] Figure 1This paper presents a schematic diagram of a computing network structure under the SRv6 protocol, which involves the coordinated operation of terminal devices, base stations, and edge gateways. The edge gateway acts as a routing node, while the edge server is a deeply integrated edge computing node with the cloud computing center. The entire system presents a multi-layered, collaborative computing network architecture, enabling efficient distribution and utilization of computing power from the terminal to the cloud. During data transmission in the multi-layered computing network system, the tasks of each component are as follows:

[0028] (1) The customer service sends a request from the terminal device and transmits it to the base station through the wireless link.

[0029] (2) In process (1), the edge gateway undertakes the task of route allocation to realize the rapid flow of data. In real network deployment, the edge gateway is usually set up close to the base station, and data transmission between each routing node is realized through real-time dynamic links.

[0030] (3) Edge servers are the core of edge applications, and they process business based on storage and computing resources.

[0031] (4) A cloud computing center is a large server cluster that is far away from customers and has massive storage and computing resources.

[0032] Elastic scheduling is an important indicator for evaluating the performance of computing networks. How to rationally allocate computing resources to complete the transmission and processing of various customer services is key to improving the user experience.

[0033] In multi-tiered computing network systems, the high-efficiency transmission of customer business data places extremely stringent demands on routing. This necessitates the proper configuration of routing nodes to ensure smooth data flow along the transmission path. To achieve this goal, the SRv6 protocol technology is used for routing scheduling in multi-tiered computing networks.

[0034] The SRv6 protocol operates similarly to computer programming. It transforms the service requirements carried by the network into instructions that can be executed by devices. These instructions are executed by network devices on the transmission path, thereby enabling the free arrangement and customization of network services.

[0035] Based on the technical characteristics and operating mechanism of Software Defined Network (SDN) and SRv6 protocol, this application proposes a routing model for a multi-layered computing network. The SDN controller is responsible for collecting and analyzing network topology information, formulating optimal route scheduling according to the individual needs of end customers, and implementing these scheduling through SRv6 protocol technology.

[0036] Specifically, this application aims to provide a service processing scheme for a computing network using the aforementioned SRv6 protocol. This scheme can configure the transmission path for sending target customer services to target computing nodes based on the characteristics of the target customer's services and the target attribute status of each routing node, thereby ensuring that the target customer's services can be more reasonably allocated to computing resources for processing.

[0037] To enable those skilled in the art to better understand the technical solutions in this specification, the technical solutions in the embodiments of this application will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of this specification, and not all embodiments. Based on the embodiments in this specification, all other embodiments obtained by those skilled in the art without creative effort should fall within the scope of protection of this specification.

[0038] One embodiment of this application proposes a business processing method. Figure 2 This is a flowchart illustrating the business processing method, including:

[0039] S202, Identify the target computing node in the computing network responsible for processing the target customer's business.

[0040] In this embodiment, the computing network includes multiple computing nodes and multiple routing nodes. The computing nodes are responsible for processing the assigned client services, and the routing nodes form the transmission paths from the terminal devices to the computing nodes. That is, after a client service is sent from a terminal device, it can be transmitted to the computing node for processing according to the transmission path formed by the corresponding routing nodes. It should be understood that the transmission path between the terminal device and each computing node is not unique. How to schedule and plan reasonable transmission paths for the routing nodes is crucial to ensuring the smooth flow of client service data.

[0041] Specifically, in this step, after the terminal device initiates the target customer service to be processed, a target computing node responsible for processing can be assigned to it according to a certain load balancing strategy.

[0042] As an example, the idle time of multiple computing nodes in a computing network can be determined based on at least one of the computing node's storage resource capacity, computing resource capacity, and data transmission rate. Then, a target computing node whose idle time meets a certain preset standard is selected from the multiple computing nodes to handle the target customer's business. For example, the computing node with the highest idle time is selected as the target computing node to handle the target customer's business.

[0043] S204: Select a preset number of baseline routing nodes from multiple routing nodes in the computing power network, taking the connectivity cost relative to the target computing node as the dimension.

[0044] In this embodiment, the connectivity cost of any target routing node relative to the target computing node is determined based on the adjacency matrix of the target routing node relative to the target computing node and / or the communication delay of the target routing node relative to the target computing node.

[0045] In one feasible implementation, the connectivity cost of each routing node relative to the target computing node in the computing power network can be determined first based on the adjacency matrix of the routing node relative to the target computing node and / or the communication delay of the routing node relative to the target computing node; then, a preset number of routing nodes whose connectivity costs meet a certain preset standard are selected as the baseline routing nodes.

[0046] The adjacency matrix describes the adjacency relationship between routing nodes, that is, whether the communication between any two routing nodes is one-hop or multi-hop. As an example, assuming there are 5 routing nodes, the corresponding adjacency matrix L can be described by the following formula:

[0047]

[0048] For the adjacency matrix L above, from the row direction, the first row is route node 1, the second row is route node 2, ..., the fifth row is route node 5. From the column direction, the first column is route node 1, the second column is route node 2, ..., the fifth column is route node 5. The diagonal represents routes between themselves, so it is represented by 0. Except for the diagonal, 1 indicates one hop, and 0 indicates two or more hops. For example, if the value in the third row and second column is 1, it means that route node 3 and route node 2 only require one hop; conversely, if the value in the fifth row and first column is 0, it means that route node 5 and route node 1 require two or more hops.

[0049] As can be seen, the adjacency matrix can clearly indicate the hops between routing nodes in computation. That is, the more hops there are, the greater the connectivity cost; the fewer hops there are, the smaller the connectivity cost.

[0050] Furthermore, the communication delay between any target routing node and the target computing node is determined based on the channel bandwidth and / or data transmission rate of the target customer service corresponding to all associated routing nodes. The associated routing node is the routing node on the shortest transmission path from the target customer service to the target computing node. For example, when calculating the communication delay between routing node 3 and the target computing node, if the shortest transmission path between the terminal device and the target computing node including routing node 3 is "routing node 5 → routing node 3 → routing node 7", then the communication delay between routing node 3 and the target computing node needs to be calculated based on the channel bandwidth and / or data transmission rate of the target customer service corresponding to routing nodes 5, 3, and 7.

[0051] As an example, this embodiment can employ Shannon's law to determine the data transmission rate of the target customer service across all associated routing nodes based on the signal-to-noise ratio (SNR) of the target customer service across all associated routing nodes. The calculation methods for the relevant parameters are described below.

[0052] According to Shannon's law, the target customer's service g is routed to the receiving routing node U. g The data transfer rate between them is described by the following formula:

[0053]

[0054] In the formula, Used to describe the target customer's business g to all associated routing nodes U g Channel bandwidth between Used to describe customer service g to all associated routing nodes U g The signal-to-noise ratio between them.

[0055] in, Used to describe the following formula

[0056]

[0057] ρ0、 These are used to describe the spectral density of additive white Gaussian noise, and the target customer service g to all associated routing nodes U, respectively. g Path consumption between them, target customer service g to all associated routing nodes U g Transmit power between [various parameters]. Target customer service g to all associated routing nodes U. g The longer the distance between them, the better. The more, the better. A normalized quantization standard is used to express data transmission rate.

[0058] From the above, we can obtain the route from the target customer service g to all associated routing nodes U. g The reception delay between them can be described by the following formula:

[0059]

[0060] Within a multi-layered computing network, target customer services transmit data between adjacent routing nodes via dynamic communication links. If the data transmission rate of the dynamic link can be detected, it can be described by the following formula:

[0061]

[0062] Among them, v i,j Used to describe the data transmission rate between two routing nodes i and j, and conforms to the following formula:

[0063]

[0064] Furthermore, the transmission delay of the target customer service g between the two routing nodes i and j is described by the following formula:

[0065]

[0066] In a multi-layered computing network, data is transmitted between computing nodes and routing nodes using fixed links. The data transmission rate between routing node i and computing node x is described by the following formula:

[0067]

[0068] Therefore, assuming that compute node x is selected by the target customer service g, the communication delay from the routing node to compute node x is described by the following formula:

[0069]

[0070] In practical applications, the appropriateness of the cluster number setting directly affects the clustering effect. Besides setting the cluster number directly according to requirements, this embodiment can also pre-test different cluster numbers and ultimately adopt the cluster number that yields the best results.

[0071] For example, using the target attribute as a dimension, multiple routing nodes can be clustered according to various cluster numbers to obtain the sample clustering results corresponding to each cluster number, and then the clustering quality coefficient of each sample clustering result can be calculated.

[0072] The clustering quality coefficient for each sample clustering result is determined based on the cohesion and separation of that sample clustering result. Cohesion characterizes the distance between routing nodes within a cluster and assesses the differences between routing nodes within the same cluster; higher cohesion indicates a worse clustering effect. Separation characterizes the distance between clusters and assesses the differences between different clusters; higher separation indicates a better clustering effect. Therefore, from multiple sample clustering results, one clustering quality coefficient is selected as the target sample clustering result in ascending order, and the final cluster number corresponding to the target sample clustering result is adopted.

[0073] As a feasible implementation method, the formula for calculating the clustering quality coefficient D is:

[0074]

[0075] Among them, D in Indicates cohesion. D in Indicates the degree of separation. P i m e is used to describe the i-th cluster, the total number of routing nodes in the multi-level computing power network, the cluster center of the i-th cluster, and the routing nodes in the i-th cluster, respectively. This is used to describe the mean distance between the cluster centers of each cluster.

[0076] It should be noted that the above formula for calculating the clustering quality coefficient D is only for illustrative purposes. Any implementation method in which the clustering quality coefficient of the sample clustering result is positively correlated with the cohesion of the sample clustering result and negatively correlated with the separation of the sample clustering result should fall within the protection scope of this specification.

[0077] It should be understood that the baseline routing node in this embodiment is a high-quality routing node with low cost of connecting to the target computing node, and subsequent transmission paths need to be configured around the quasi-routing node.

[0078] S206. Each base routing node is used as the initial cluster center. The target attribute of the routing node is used as the dimension to cluster all non-base routing nodes in multiple routing nodes to obtain a preset number of clusters. Different clusters correspond to different value ranges of the target attribute.

[0079] The target attributes in this embodiment can be, but are not limited to, service priority, hardware specifications, energy consumption, current traffic, etc.

[0080] Taking hardware specifications as an example, clustering all non-baseline routing nodes among multiple routing nodes according to hardware specifications yields clusters for different hardware specification value ranges. For instance, by setting the hardware specification value ranges to high, medium, and low hardware specifications, we can classify which routing nodes have high hardware specifications, which have medium hardware specifications, and which have low hardware specifications.

[0081] Similarly, taking energy consumption as an example, clustering all non-baseline routing nodes among multiple routing nodes according to energy consumption can yield clusters for different energy consumption ranges. For instance, by setting the energy consumption range to high, medium, and low energy consumption, we can classify which routing nodes are high-energy-consuming, which are medium-energy-consuming, and which are low-energy-consuming.

[0082] The clustering process involves determining the required number of clusters and selecting routing nodes matching the desired number as the initial cluster centers. Then, using the target attribute as the dimension, the mathematical distance between all non-baseline routing nodes and each cluster center is calculated, and the routing nodes are added to the cluster with the smallest mathematical distance (or a mathematical distance less than a standard). It should be understood that adding non-baseline routing nodes to each cluster will change the cluster center. After updating the cluster centers, the mathematical distance between all non-baseline routing nodes and each cluster center is recalculated using the target attribute as the dimension for classification. This process is iterated repeatedly until the cluster centers no longer change or the changes reach a certain convergence condition. It should be noted that this embodiment only clusters non-baseline routing nodes, while the baseline routing nodes in each cluster remain unchanged.

[0083] S208: From a preset number of clusters, determine the target cluster whose target attribute value range matches the target customer service, and select at least two routing nodes from the target cluster, including the corresponding baseline routing node, to configure the transmission path for sending the target customer service to the target computing node.

[0084] In this embodiment, each cluster corresponds to a range of values ​​for a target attribute.

[0085] As an example, assuming the target attribute is service priority, the resulting clusters are divided into high service priority clusters, medium service priority clusters, and low service priority clusters. If the target customer service is a high-priority service, at least two routing nodes, including the base routing node, can be selected from the high service priority cluster to configure the transmission path.

[0086] Alternatively, assuming the target attribute is energy consumption, the resulting clusters are divided into high-energy-consumption clusters, medium-energy-consumption clusters, and low-energy-consumption clusters. If the target customer's business adopts an energy-saving strategy, at least two routing nodes, including the base routing node, can be selected from the low-energy-consumption cluster to configure the transmission path.

[0087] It should be understood that the routing nodes selected for configuring the transmission path must include a base routing node, and the base routing node is a high-quality routing node with low cost of connecting to the target computing node. Therefore, the overall cost of the transmission path configured around the base routing node is also relatively small.

[0088] S210: Send the target customer service to the target computing node for service processing according to the transmission path.

[0089] In summary, the method of this embodiment, when a target customer service needs to be processed by the computing power network, firstly determines the target computing node responsible for processing the target customer service, and selects a preset number of baseline routing nodes from multiple routing nodes in the computing power network, using the connection cost relative to the target computing node as a dimension. Next, each baseline routing node is used as an initial cluster center, and all non-baseline routing nodes among the multiple routing nodes are clustered using the target attribute of the routing node as a dimension, resulting in clusters with different value ranges of the target attribute. Finally, based on the characteristics of the target customer service, a target cluster with the most matching value range of the target attribute is determined, and at least two routing nodes, including the corresponding baseline routing node, are selected from the target cluster to configure the transmission path for sending the target customer service to the target computing node. Since the connection cost of the baseline routing node relative to the target computing node is relatively small, further selecting other non-baseline routing nodes with similar target attributes around the baseline routing node can configure a transmission path with a relatively low overall cost, thereby making more rational use of computing power network resources and allocating the target customer service to computing nodes for processing.

[0090] The method of this embodiment will be explained below with reference to a practical application scenario.

[0091] This application scenario is used to balance the load of routing nodes in a large shopping mall. To this end, after identifying a certain number of baseline routing nodes, the remaining non-baseline routing nodes in the large shopping mall are clustered based on current traffic. This identifies clusters with different current traffic ranges (i.e., current load). Then, a baseline routing node and a certain number of other non-baseline routing nodes are selected from the cluster with the smallest current traffic range to handle customer traffic.

[0092] During the solution verification process, the configuration of the routing nodes (routers) in the large shopping mall is shown in the table below:

[0093]

[0094] Four different routing nodes (A, B, C, and D) were selected in the shopping mall, and the traffic data for every 15 minutes is shown in the table below:

[0095]

[0096]

[0097] The data in the table above shows that the traffic of routing node B is significantly higher, while the traffic of node D, although lower than that of B, is still far higher than that of nodes A and C, and the traffic of node A is relatively the lowest. This significant uneven distribution of traffic not only limits the network performance before routing scheduling but also leads to a waste of resources.

[0098] After using the scheme of the embodiment to schedule routing nodes to handle customer services, the traffic data of each routing node every 15 minutes is shown in the table below:

[0099]

[0100] As shown in the table above, after scheduling using the method described in this embodiment, the traffic of each routing node has become nearly balanced, with no significant differences in traffic volume. This indicates that the traffic distribution has become more uniform after routing node scheduling. This not only improves the performance and stability of the computing network but also increases resource utilization.

[0101] Corresponding to Figure 2 The method shown in this application is another embodiment of a service processing apparatus applied to a computing network. Figure 3 This is a schematic diagram of the structure of the service processing device 300, including:

[0102] The computing node configuration module 310 is used to determine the target computing node in the computing power network that is responsible for processing the target customer's business.

[0103] Clustering preparation module 320 is used to select a preset number of reference routing nodes from multiple routing nodes of the computing power network, with the connectivity cost relative to the target computing node as the dimension.

[0104] The clustering execution module 330 is used to take each of the base routing nodes as the initial cluster center, and to cluster all non-base routing nodes of the plurality of routing nodes with the target attribute of the routing node as the dimension, so as to obtain the preset number of clusters, and different clusters correspond to different value ranges of the target attribute.

[0105] The path determination module 340 is used to determine, from the preset number of clusters, a target cluster whose value range of the target attribute matches the target customer service, and select from the target cluster at least two routing nodes including the corresponding baseline routing node, so as to configure a transmission path for sending the target customer service to the target computing node.

[0106] The service sending module 350 is used to send the target customer service to the target computing node for service processing according to the transmission path.

[0107] In this embodiment, when a target customer service needs to be processed by the computing network, the apparatus first determines the target computing node responsible for processing the target customer service. Then, based on the connectivity cost relative to the target computing node, a predetermined number of baseline routing nodes are selected from multiple routing nodes in the computing network. Next, each baseline routing node is used as an initial cluster center. Based on the target attribute of the routing node, all non-baseline routing nodes among the multiple routing nodes are clustered to obtain clusters with different value ranges of the target attribute. Finally, considering the characteristics of the target customer service, a target cluster with the most matching value range of the target attribute is determined. At least two routing nodes, including the corresponding baseline routing node, are selected from this target cluster to configure the transmission path for sending the target customer service to the target computing node. Since the connectivity cost of the baseline routing node relative to the target computing node is relatively small, further selecting other non-baseline routing nodes with similar target attributes around the baseline routing node can configure a transmission path with a relatively low overall cost, thereby making more rational use of computing network resources and allocating the target customer service to computing nodes for processing.

[0108] The connectivity cost of any target routing node relative to the target computing node among the plurality of routing nodes is determined based on the adjacency matrix of the target routing node relative to the target computing node and / or the communication delay of the target routing node relative to the target computing node.

[0109] Optionally, the communication delay between the target routing node and the target computing node is determined based on the channel bandwidth and / or data transmission rate of the target customer service corresponding to all associated routing nodes; the associated routing node is the routing node on the shortest transmission path from the target customer service to the target computing node via the target routing node.

[0110] Optionally, the data transmission rate of the target customer service corresponding to all associated routing nodes is determined based on the signal-to-noise ratio of the target customer service corresponding to all associated routing nodes.

[0111] Optionally, the preset quantity is determined based on the number of clusters corresponding to the target sample clustering results; the target sample clustering results are selected from multiple sample clustering results in ascending order of clustering merit coefficients; the multiple sample clustering results are obtained by clustering multiple routing nodes according to multiple cluster numbers with the target attribute as the dimension; the clustering merit coefficient of each sample clustering result is positively correlated with the cohesion corresponding to the sample clustering result and negatively correlated with the separation corresponding to the sample clustering result; the cohesion is used to characterize the distance between routing nodes within a cluster; the separation is used to characterize the distance between clusters.

[0112] Optionally, the computing node configuration module 310 determines the target computing node in the computing power network responsible for processing the target customer's business, including: determining the idle degree of multiple computing nodes in the computing power network based on at least one of the computing node's storage resource capacity, computing resource capacity, and data transmission rate; and selecting a target computing node from the multiple computing nodes whose idle degree reaches a second preset standard to be responsible for processing the target customer's business.

[0113] It should be noted that the service processing device in this embodiment can be used as... Figure 1 The execution body of the method shown is therefore able to achieve... Figure 1 The steps and functions of the method shown will not be repeated here.

[0114] Figure 4 This is a schematic diagram of the structure of an electronic device provided in one embodiment of this application. Please refer to it. Figure 4 At the hardware level, the electronic device includes a processor, and optionally also includes an internal bus, a network interface, and memory. The memory may include main memory, such as high-speed random-access memory (RAM), or non-volatile memory, such as at least one disk drive. Of course, the electronic device may also include other hardware required for other business operations.

[0115] The processor, network interface, and memory can be interconnected via an internal bus, which can be an ISA (Industry Standard Architecture) bus, a PCI (Peripheral Component Interconnect) bus, or an EISA (Extended Industry Standard Architecture) bus, etc. This bus can be divided into address bus, data bus, control bus, etc. For ease of representation, Figure 4 The symbol is represented by a single double-headed arrow, but this does not mean that there is only one bus or one type of bus.

[0116] Memory is used to store programs. Specifically, programs may include program code, which includes computer operation instructions. Memory may include main memory and non-volatile memory, and provides instructions and data to the processor.

[0117] The processor reads the corresponding computer program from non-volatile memory into main memory and then executes it, forming the aforementioned business processing device at the logical level. The processor executes the program stored in memory and specifically performs the following operations:

[0118] Identify the target computing nodes in the computing network responsible for processing the target customer's business.

[0119] Using the connectivity cost relative to the target computing node as a dimension, a preset number of reference routing nodes are selected from multiple routing nodes in the computing power network.

[0120] Each of the baseline routing nodes is used as the initial cluster center. All non-baseline routing nodes among the multiple routing nodes are clustered using the target attribute of the routing node as the dimension to obtain the preset number of clusters. Different clusters correspond to different value ranges of the target attribute.

[0121] From the preset number of clusters, a target cluster whose value range of the target attribute matches the target customer service is determined, and at least two routing nodes, including the corresponding baseline routing node, are selected from the target clusters to configure the transmission path for sending the target customer service to the target computing node.

[0122] The target customer service is sent to the target computing node for service processing according to the transmission path.

[0123] The above is as stated in this application. Figure 2The methods disclosed in the illustrated embodiments can be applied to or implemented by a processor. The processor may be an integrated circuit chip with signal processing capabilities. During implementation, each step of the above methods can be completed by integrated logic circuits in the processor's hardware or by instructions in software form. The processor can be a general-purpose processor, including a Central Processing Unit (CPU), a Network Processor (NP), etc.; it can also be a Digital Signal Processor (DSP), an Application Specific Integrated Circuit (ASIC), a Field-Programmable Gate Array (FPGA), or other programmable logic devices, discrete gate or transistor logic devices, or discrete hardware components. It can implement or execute the methods, steps, and logic block diagrams disclosed in one or more embodiments of this application. The general-purpose processor can be a microprocessor or any conventional processor. The steps of the methods disclosed in one or more embodiments of this application can be directly implemented by a hardware decoding processor, or by a combination of hardware and software modules in the decoding processor. The software module can reside in a mature storage medium in the field, such as random access memory, flash memory, read-only memory, programmable read-only memory, electrically erasable programmable memory, or registers. This storage medium is located in memory, and the processor reads information from the memory and, in conjunction with its hardware, completes the steps of the above method.

[0124] Of course, in addition to software implementation, the electronic device of this application does not exclude other implementation methods, such as logic devices or a combination of hardware and software, etc. In other words, the execution subject of the following processing flow is not limited to each logic unit, but can also be hardware or logic devices.

[0125] This application also proposes a computer program product, which includes a computer-readable storage medium storing a computer program operable to cause a computer to perform the following operations:

[0126] Identify the target computing nodes in the computing network responsible for processing the target customer's business.

[0127] Using the connectivity cost relative to the target computing node as a dimension, a preset number of reference routing nodes are selected from multiple routing nodes in the computing power network.

[0128] Each of the baseline routing nodes is used as the initial cluster center. All non-baseline routing nodes among the multiple routing nodes are clustered using the target attribute of the routing node as the dimension to obtain the preset number of clusters. Different clusters correspond to different value ranges of the target attribute.

[0129] From the preset number of clusters, a target cluster whose value range of the target attribute matches the target customer service is determined, and at least two routing nodes, including the corresponding baseline routing node, are selected from the target clusters to configure the transmission path for sending the target customer service to the target computing node.

[0130] The target customer service is sent to the target computing node for service processing according to the transmission path.

[0131] In summary, the above description is merely a preferred embodiment of this application and is not intended to limit the scope of protection of this application. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of one or more embodiments of this application should be included within the scope of protection of one or more embodiments of this application.

[0132] The systems, devices, and modules described in the above embodiments can be implemented by computer chips or physical entities, or by products with certain functions. A typical implementation device is a computer. Specifically, a computer can be, for example, a personal computer, laptop computer, cellular phone, camera phone, smartphone, personal digital assistant, media player, navigation device, email device, game console, tablet computer, wearable device, or any combination of these devices.

[0133] Computer-readable media includes both permanent and non-permanent, removable and non-removable media that can store information using any method or technology. Information can be computer-readable instructions, data structures, modules of programs, or other data. Examples of computer storage media include, but are not limited to, phase-change memory (PRAM), static random access memory (SRAM), dynamic random access memory (DRAM), other types of random access memory (RAM), read-only memory (ROM), electrically erasable programmable read-only memory (EEPROM), flash memory or other memory technologies, CD-ROM, digital versatile optical disc (DVD) or other optical storage, magnetic tape, magnetic disk storage or other magnetic storage devices, or any other non-transferable medium that can be used to store information accessible by a computing device. As defined herein, computer-readable media does not include transient computer-readable media, such as modulated data signals and carrier waves.

[0134] It should also be noted that the terms "comprising," "including," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such a process, method, article, or apparatus. Without further limitation, an element defined by the phrase "comprising one..." does not exclude the presence of other identical elements in the process, method, article, or apparatus that includes said element.

[0135] The various embodiments in this application are described in a progressive manner. Similar or identical parts between embodiments can be referred to mutually. Each embodiment focuses on describing the differences from other embodiments. In particular, the system embodiments are basically similar to the method embodiments, so the description is relatively simple; relevant parts can be referred to the descriptions of the method embodiments.

Claims

1. A business processing method applied to a computing power network, characterized in that, The method includes: Identify the target computing nodes in the computing network responsible for processing the target customer's business; Using the connectivity cost relative to the target computing node as a dimension, a preset number of reference routing nodes are selected from multiple routing nodes of the computing power network. Each of the baseline routing nodes is used as the initial cluster center. All non-baseline routing nodes among the multiple routing nodes are clustered using the target attribute of the routing node as the dimension to obtain the preset number of clusters. Different clusters correspond to different value ranges of the target attribute. From the preset number of clusters, determine the target cluster whose value range of the target attribute matches the target customer service, and select at least two routing nodes from the target cluster, including the corresponding baseline routing node, to configure the transmission path for sending the target customer service to the target computing node. The target customer service is sent to the target computing node for service processing according to the transmission path.

2. The method according to claim 1, characterized in that, The connectivity cost of any target routing node relative to the target computing node among the plurality of routing nodes is determined based on the adjacency matrix of the target routing node relative to the target computing node and / or the communication delay of the target routing node relative to the target computing node.

3. The method according to claim 2, characterized in that, The communication latency between the target routing node and the target computing node is determined based on the channel bandwidth and / or data transmission rate of the target customer service corresponding to all associated routing nodes; the associated routing node is the routing node on the shortest transmission path from the target customer service to the target computing node via the target routing node.

4. The method according to claim 3, characterized in that, The data transmission rate of the target customer service corresponding to all associated routing nodes is determined based on the signal-to-noise ratio of the target customer service corresponding to all associated routing nodes.

5. The method according to any one of claims 1 to 4, characterized in that, The preset quantity is determined based on the number of clusters corresponding to the target sample clustering results; the target sample clustering results are selected from multiple sample clustering results in ascending order of clustering merit coefficients; the multiple sample clustering results are obtained by clustering multiple routing nodes according to multiple clustering numbers with the target attribute as the dimension; the clustering merit coefficient of each sample clustering result is positively correlated with the cohesion corresponding to the sample clustering result and negatively correlated with the separation corresponding to the sample clustering result; the cohesion is used to characterize the distance between routing nodes within a cluster; the separation is used to characterize the distance between clusters.

6. The method according to any one of claims 1 to 4, characterized in that, Identify the target computing nodes in the computing network responsible for processing the target customer's business, including: The idle time of multiple computing nodes in the computing power network is determined based on at least one of the storage resource capacity, computing resource capacity and data transmission rate of the computing nodes. Select one target computing node from multiple computing nodes that meets the second preset standard for idle time to handle the target customer's business.

7. The method according to any one of claims 1 to 4, characterized in that, The target attributes include at least one of the following: business priority, hardware specifications, and energy consumption.

8. A service processing device, applied to a computing network, characterized in that, include: The compute node configuration module is used to determine the target compute nodes in the computing network that are responsible for processing the target customer's business. The clustering preparation module is used to select a preset number of baseline routing nodes from multiple routing nodes of the computing power network, with the connectivity cost relative to the target computing node as the dimension. The clustering execution module is used to take each of the baseline routing nodes as the initial cluster center, and to cluster all non-baseline routing nodes of the plurality of routing nodes with the target attribute of the routing node as the dimension, so as to obtain the preset number of clusters, and different clusters correspond to different value ranges of the target attribute. The path determination module is used to determine, from the preset number of clusters, the target cluster whose value range of the target attribute matches the target customer service, and select from the target cluster at least two routing nodes including the corresponding baseline routing node, so as to configure the transmission path for sending the target customer service to the target computing node. The service sending module is used to send the target customer service to the target computing node for service processing according to the transmission path.

9. An electronic device, comprising: processor; And a memory arranged to store computer-executable instructions, which, when executed, cause the processor to perform the method as described in any one of claims 1 to 7.

10. A computer program product comprising a computer-readable storage medium storing a computer program operable to cause a computer to perform the method as claimed in any one of claims 1 to 7.

Images