CN117278566A - Computing power node selection method, device, electronic equipment and readable storage medium - Google Patents

Abstract

Embodiments of this specification disclose a computing power node selection method, device, electronic equipment and readable storage medium. The method includes: obtaining multiple index data of each computing power node in the computing power network to determine the performance of each computing power node. Rank sum ratio set; obtain the data transmission delay between the requesting node and other computing power nodes in the computing power network; the requesting node is the node that issues the calculation request among each computing power node; according to the preset initial value of pheromone concentration, The rank sum ratio set and the data transmission delay between the requesting node and other computing nodes in the computing power network are used to determine the pheromone concentration on the path from the requesting node to other computing power nodes; based on the relationship between the requesting node and other computing power nodes in the computing power network The data transmission delay between power nodes and the pheromone concentration on the path from the requesting node to other computing power nodes are used to determine the target computing power node that will provide computing services to the requesting node.

Description

Computing power node selection method, device, electronic equipment and readable storage medium

Technical field

This document relates to the field of computing resource scheduling, and in particular to a computing node selection method, device, electronic equipment and readable storage media.

Background technique

With the deep integration of cloud computing and networks, computing power will become the core productivity in the next-generation operational service model. Massive edge computing nodes, supercomputing centers and "Eastern and Western computing" hub nodes are all important components of the computing power network. With the help of communication infrastructure such as 5G and industrial Internet, computing resources can be allocated and flexibly scheduled on demand. In response to widely distributed user computing requests, how to select computing power nodes and scheduling paths to provide users with better computing services to obtain an overall better user experience is an urgent technical issue that needs to be solved.

Contents of the invention

The purpose of the embodiments of this specification is to provide a computing power node selection method, device, electronic device and readable storage medium for determining the optimal computing power node and the optimal scheduling path in the computing power network.

In order to solve the above technical problems, the embodiments of this specification are implemented as follows:

In the first aspect, a computing power node selection method is proposed, including:

Obtain multiple index data of each computing power node in the computing power network to determine the rank sum ratio set of each computing power node, where the multiple index data of the computing power node is used to characterize the service capability of the computing power node ;

Obtain the data transmission delay between the requesting node and other computing power nodes in the computing power network; the requesting node is the node that issues the computing request among the computing power nodes;

According to the preset initial value of pheromone concentration, the rank sum ratio set and the data transmission delay between the requesting node and other computing power nodes in the computing power network, determine the distance between the requesting node and other computing power nodes. Pheromone concentration along the path of the node;

According to the data transmission delay between the requesting node and other computing power nodes in the computing power network, and the pheromone concentration on the path from the requesting node to other computing power nodes, it is determined that the requesting node will be provided with The target computing power node of the computing service.

In the second aspect, a computing power node selection device is proposed, including:

The data acquisition unit acquires multiple index data of each computing power node in the computing power network to determine the rank sum ratio set of each computing power node, wherein the multiple index data of the computing power node is used to characterize the computing power. The service capability of the node;

A delay acquisition unit obtains the data transmission delay between the requesting node and other computing nodes in the computing power network; the requesting node is the node that issues the computing request among the computing power nodes;

The processing unit determines the arrival of the requesting node based on the preset initial value of the pheromone concentration, the rank sum ratio set, and the data transmission delay between the requesting node and other computing power nodes in the computing power network. Pheromone concentration on the path of other computing power nodes;

The determining unit determines the upcoming data transmission time between the requesting node and other computing power nodes in the computing power network and the pheromone concentration on the path from the requesting node to other computing power nodes. The target computing power node that requests the node to provide computing services.

In the third aspect, an electronic device is proposed, including:

processor; and

Memory arranged to store computer-executable instructions that, when executed, cause the processor to:

Obtain multiple index data of each computing power node in the computing power network to determine the rank sum ratio set of each computing power node, where the multiple index data of the computing power node is used to characterize the service capability of the computing power node ;

Obtain the data transmission delay between the requesting node and other computing power nodes in the computing power network; the requesting node is the node that issues the computing request among the computing power nodes;

According to the preset initial value of pheromone concentration, the rank sum ratio set and the data transmission delay between the requesting node and other computing power nodes in the computing power network, determine the distance between the requesting node and other computing power nodes. Pheromone concentration along the path of the node;

According to the data transmission delay between the requesting node and other computing power nodes in the computing power network, and the pheromone concentration on the path from the requesting node to other computing power nodes, it is determined that the requesting node will be provided with The target computing power node of the computing service.

A fourth aspect proposes a computer-readable storage medium that stores one or more programs that, when executed by an electronic device including a plurality of application programs, cause the The electronic device described above performs the following operations:

Obtain multiple index data of each computing power node in the computing power network to determine the rank sum ratio set of each computing power node, where the multiple index data of the computing power node is used to characterize the service capability of the computing power node ;

Obtain the data transmission delay between the requesting node and other computing power nodes in the computing power network; the requesting node is the node that issues the computing request among the computing power nodes;

According to the preset initial value of pheromone concentration, the rank sum ratio set and the data transmission delay between the requesting node and other computing power nodes in the computing power network, determine the distance between the requesting node and other computing power nodes. Pheromone concentration along the path of the node;

According to the data transmission delay between the requesting node and other computing power nodes in the computing power network, and the pheromone concentration on the path from the requesting node to other computing power nodes, it is determined that the requesting node will be provided with The target computing power node of the computing service.

It can be seen from the technical solutions provided by the above embodiments of this specification that the embodiments of this specification have at least one of the following technical effects:

Obtain multiple indicator data of each computing power node in the computing power network to determine the rank sum ratio set of each computing power node. Among them, multiple indicator data of the computing power node are used to characterize the service capabilities of the computing power node; obtain the request node The data transmission delay with other computing power nodes in the computing power network; the requesting node is the node that issues the calculation request among each computing power node; according to the preset initial value of pheromone concentration, the rank sum ratio set and the request node and The data transmission delay between other computing power nodes in the computing power network determines the pheromone concentration on the path from the requesting node to other computing power nodes; based on the data transmission time between the requesting node and other computing power nodes in the computing power network Delay, and the pheromone concentration on the path from the requesting node to other computing power nodes, determine the target computing power node that will provide computing services to the requesting node.

It can obtain multiple indicator data of each computing power node in the computing power network, determine the rank sum ratio set of computing power nodes, and obtain the data transmission delay between the requesting node and other computing power nodes. Based on these data and preset The initial value of the pheromone concentration can determine the best path from the requesting node to other computing nodes, and find the target computing node that is most suitable for providing computing services. This optimization process can be used under the limited computing resources and bandwidth of the computing system. Under the conditions, the user response delay is reduced, the efficiency and performance of the computing network are improved, and the requesting node receives high-quality computing services.

Description of the drawings

In order to more clearly explain the embodiments of this specification or the technical solutions in the prior art, the drawings needed to be used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings in the following description are only These are some of the embodiments recorded in this specification. For those of ordinary skill in the art, other drawings can be obtained based on these drawings without exerting any creative effort.

Figure 1 is a schematic diagram of the overall architecture of a computing power network provided by an embodiment of this specification.

Figure 2 is a schematic flowchart of the implementation of a computing power node selection method provided by an embodiment of this specification.

Figure 3 is a schematic flowchart of the implementation of another computing power node selection method provided by an embodiment of this specification.

Figure 4 is a schematic structural diagram of a computing power node selection device provided by an embodiment of this specification.

FIG. 5 is a schematic structural diagram of an electronic device according to an embodiment of the specification.

Detailed ways

In order to make the purpose, technical solutions and advantages of this document clearer, the technical solutions of this specification will be clearly and completely described below in conjunction with specific embodiments of this specification and the corresponding drawings. Obviously, the described embodiments are only some of the embodiments of this specification, but not all of the embodiments. Based on the embodiments in this specification, all other embodiments obtained by those of ordinary skill in the art without creative efforts fall within the scope of protection of this document.

The technical solutions provided by each embodiment of this specification will be described in detail below with reference to the accompanying drawings.

The computing power node selection method provided by this invention is applied in the overall architecture of the computing power network to select the optimal scheduling path and the target computing power node used to provide computing services for massive user requests at any location to ensure that Under the conditions of limited computing power service nodes and bandwidth resources, obtain the overall optimal user experience.

Please refer to Figure 1. Figure 1 is a schematic diagram of the overall architecture of a computing power network provided by an embodiment of this specification. As shown in Figure 1, the computing power network includes a four-layer architecture of control layer 10, computing power service layer 20, network layer 30 and user layer 40. Among them, the control layer 10 is mainly a centralized control center 11, which controls the normal operation of the entire computing system; the computing service layer 20 is mainly composed of a supercomputing center 21, edge computing nodes 23, etc., and mainly provides computing power and storage services to users. , the supercomputing center 21 is generally deployed far away from the user, with sufficient storage and computing resources, and the edge computing node 23 is close to the user, but has limited storage and computing resources; the network layer 30 mainly includes various network devices 33, the network The device 33 is used to provide data transmission services between users and computing power nodes; the user layer 40 includes various users 43 with computing needs. When the user 43 initiates a computing request, the optimal target computing power node is selected to provide computing services through the computing power node selection method provided in the present invention, and the shortest data transmission path to the target computing power node is determined.

In the computing power node selection method provided by the embodiments of this specification, the execution subject of the method is executed by a computer device, for example, by at least one of a server, a laptop computer, a desktop computer, a tablet computer, or an intelligent robot. Or it can also be a centralized control center in the computing power network shown in Figure 1.

For the convenience of description, the following takes the execution subject of this method as an electronic device capable of executing the computing power node selection method as an example. The electronic device can specifically be an electronic device such as a server, a laptop computer, a desktop computer, a tablet computer, or an intelligent robot. The implementation of this method is introduced. It can be understood that the fact that the execution subject of the method is an electronic device is only an exemplary description and should not be understood as a limitation of the method.

Figure 2 is a schematic flowchart of the implementation of a computing power node selection method provided by an embodiment of this specification, including:

S210: Obtain multiple index data of each computing power node in the computing power network to determine the rank sum ratio set of each computing power node.

As shown in Figure 1, each computing power node in the computing power network can be represented by {C ₁ , C ₂ ,..., C _n }, n is the number of computing power nodes in the computing power network, and the number of computing power nodes is Indicator data is used to characterize the service capabilities of computing nodes, including but not limited to CPU utilization, memory utilization, disk utilization and number of connections, respectively. express.

Among them, CPU utilization refers to the usage rate of the CPU processor on the computing node, indicating the ratio of the time the CPU is executing computing tasks to the total time; memory utilization refers to the usage rate of the memory on the computing node, reflecting the usage of the memory. The proportion occupied by data and programs; disk utilization refers to the usage of disk storage space on the computing node, showing the ratio between the used space and the total space in the disk; the number of connections refers to the connection between the computing node and other devices or The number of connections established between nodes, such as the number of network connections or database connections. The number of connections reflects the communication activities between computing nodes and other devices. It is very important for performance monitoring and optimization of network communication and data transmission.

These indicator data can help administrators monitor the health status of computing nodes and ensure that computing nodes maintain efficient operation during work.

The rank sum ratio method (RSR) is used to comprehensively evaluate the above-mentioned multi-dimensional indicator data to obtain the rank sum ratio set of n computing power nodes. The rank sum ratio represents the service capability of the computing power node.

S220: Obtain the data transmission delay between the requesting node and other computing power nodes in the computing power network.

Among them, the requesting node is the node that issues the computing request among the n computing power nodes.

The rank sum ratio method is used to rank the service capabilities of the computing power nodes. However, in actual applications, the computing power node with the best performance is not always selected to provide computing services. Therefore, in this step, the link is comprehensively considered. Factors such as transmission delay, geographical location, and congestion are used to form a delay matrix based on the data transmission delay between the requesting node and other computing nodes in the computing network. The minimum delay sum between computing nodes is calculated based on the delay matrix. shortest path.

S230: Determine the information on the path from the requesting node to other computing nodes based on the preset initial value of pheromone concentration, the set of rank sum ratios, and the data transmission delay between the requesting node and other computing nodes in the computing network. factor concentration.

The concentration of pheromone on a path reflects the frequency or number of times the path is used to transmit computing power requests. As time goes by, the pheromone concentration on each path is constantly updated, and the optimal target computing power node for providing services can be determined by selecting the path with the highest pheromone concentration.

S240: Determine the target computing power that will provide computing services to the requesting node based on the data transmission delay between the requesting node and other computing power nodes in the computing power network and the pheromone concentration on the path from the requesting node to other computing power nodes. node.

According to the resource utilization of the computing node server equipment in the computing network, and based on the improved rank sum ratio algorithm, a comprehensive evaluation method of computing node service capabilities is proposed. Then a data transmission delay matrix between routing nodes is established, and the shortest path between computing nodes is calculated based on this matrix. Finally, the computing power node service capabilities, latency and other indicators are reflected through the increase or decrease of pheromone concentration, adapting the user to the optimal computing power service node, and determining the shortest path, which can use the limited computing power resources of the computing power system. Under the conditions of bandwidth and bandwidth, the user response delay is reduced and the overall optimal user experience is obtained.

Figure 3 is a schematic flowchart of the implementation of another computing power node selection method provided by an embodiment of this specification. The computing power node selection method shown in Figure 3 is consistent with Figure 2. There are areas not explained in the embodiment shown in Figure 3. Reference may be made to the description in Figure 2, which will not be described again here. The computing power node selection methods shown in Figure 3 include:

S310: Obtain multiple index data of each computing power node in the computing power network to determine the rank sum ratio set of each computing power node.

Since different indicator data have different impacts on node service capabilities, in order to reasonably evaluate the service capabilities of computing power nodes, the rank sum ratio (RSR) method is used to comprehensively evaluate multi-dimensional indicator data. Illustratively, n computing power nodes and the 4 indicator data in the above example (not limited to 4, it can be any number) form an original indicator data matrix D with n rows and 4 columns:

Among them, d ₁₁ is the value of the first indicator data of the first computing power node, d ₂₁ is the value of the first indicator data of the second computing power node, and so on, which will not be repeated here.

Based on the above original indicator data matrix D, the rank data matrix R composed of the ranks of each computing power node can be obtained:

Among them, r ₁₁ is the rank value of the first indicator data of the first computing power node, r ₂₁ is the rank value of the first indicator data of the second computing power node, and so on, which will not be repeated here. Repeat.

Indicator data includes low-quality indicator data and high-quality indicator data. For low-quality indicator data, the lower the value, the better the performance of the computing power node; for high-quality indicator data, the higher the value, the better the performance of the computing power node, for example , CPU utilization, memory utilization, disk utilization, and billing indicators are all low-quality indicators. When calculating the rank value, the corresponding calculation formula can be selected based on the high-quality or low-quality attributes of the indicator data.

The calculation formula for the rank of low-quality index data includes:

The calculation formula for the rank of high-quality index data includes:

Among them, r _ij is the rank value of indicator data j of computing power node i, n is the total number of computing power nodes, d _max is the maximum value of indicator data j of n computing power nodes, d _max =max(d _1j , d _2j ,…,d _nj ,); Similarly, d _min is the minimum value of the indicator data j of n computing power nodes, d _1j is the value of the indicator data j of the first computing power node, and so on. , which will not be described in detail here.

After obtaining the rank data matrix R, the rank sum ratio (RSR) of the computing power node C _i can be calculated based on the rank data matrix R. The calculation formula of RSR includes:

Among them, RSR _i is the rank sum ratio of computing power node i, n is the total number of computing power nodes, and r _ij is the rank value of index data j of computing power node i. Finally, the set of rank sum ratios RSR ₁ , RSR ₂ ,..., RSR _n of the computing power nodes {C ₁ , C ₂ ,..., C _n } is obtained. The rank sum ratio represents the service capability of the computing power node. Therefore, by comparing the ranks Sorting the sum-ratio set can select the computing power node with the best service capability.

When each indicator data has a different impact on the performance of the computing node, the rank sum ratio of each indicator data can be given different weights as needed. Then the calculation formula of the weighted rank sum ratio is as follows:

Among them, w _j is the rank sum ratio weight of indicator data j, and different performance requirements can be met through the weighted rank sum ratio.

S320: Obtain the data transmission delay between the requesting node and other computing power nodes in the computing power network.

Among them, the requesting node is the node that issues the calculation request among the n computing power nodes, and the other computing power nodes are the computing power nodes in the computing power network except the requesting node.

The rank sum ratio method is used to rank the service capabilities of the computing power nodes. However, in actual applications, the computing power node with the best performance is not always selected to provide computing services. In this step, link transmission is comprehensively considered. Factors such as delay, geographical location, and congestion are used to calculate the minimum delay and shortest path between computing nodes based on the delay matrix.

The data transmission delay between routing nodes on the path between computing nodes in the computing network can be represented by the delay matrix T:

Among them, t _i,j represents the data transmission delay between routing nodes i, j, 1≤i≤n, 1≤j≤n, and n is the routing node. The value method of t _i,j is as follows:

in Represents the ratio of the link bandwidth and transmission rate of the path between routing nodes i and j. When a user under routing node i initiates a computing request, a computing power node under routing node j needs to be selected to provide services for it.

Calculating the minimum delay and shortest path between computing nodes based on the delay matrix T includes: determining the distance set S = [t _{i ,1} ,t _i,2 ,…,t _i,n ]; select an intermediate node m that meets the preset requirements from the distance set S; perform the distance update step based on the intermediate node m; the distance update step includes: calculating the request node correspondence The routing node i passes through the intermediate node m to the corresponding routing node j of other computing power nodes in the computing power network. The first distance set is compared with the distance in the distance set S. If in the first distance set If the distance between the corresponding routing nodes is less than the distance in the distance set S, then the distance value in the first distance set is used to update the distance value in the distance set S; then the next intermediate point that meets the preset requirements is selected from the distance set S Node m, and repeat the distance update step until the distance values in the distance set S meet the preset convergence conditions, and the target distance set is obtained.

Among them, the routing node i corresponding to the requesting node refers to the access routing node of the requesting node, that is, the user requests computing services from the computing power network through the access routing node, which can also be said to be the routing node closest to the requesting node in the communication link. . The intermediate node m that meets the preset requirements refers to the routing node whose distance is not 0. The preset convergence condition is that the distances in the distance set S are all smaller than the distance values in the first distance set. Therefore, the distances in the target distance set are the shortest distance S _short and the optimal path from routing node i to each routing node.

For example, find the array S=[t _i,1 ,t _i,2 ,...,t _i,n ] corresponding to the routing node i from the delay matrix T, and find the value that is not 0 from the array S, as The intermediate node is assumed to be ti _,m . Assume ti _,m is the shortest distance from routing node i to routing node m. The path from routing node i to routing node is recorded. Find the array corresponding to m from the delay matrix T, calculate the distance from routing node i via routing node m to other routing nodes, and compare it with the distance value in the array S, for example, from routing node i via routing node m The distance to routing node 1 is compared with t _i,1 , and the distance from routing node i to routing node 2 via routing node m is compared with t _i,2 ; if there is a path smaller than the value in array S, update S distance value and record the path. Select the next intermediate node from the array S and repeat the distance update step until the array S is no longer updated, that is, the distances in S are all smaller than the distance values in the first distance set, thereby obtaining the shortest distance from routing node i to each routing node. The distance S _short and the target distance set of the optimal path.

S330: Determine the request based on the preset computing power node charging method, the preset initial value of pheromone concentration, the rank sum ratio set, and the data transmission delay between the requesting node and other computing power nodes in the computing power network. Pheromone concentration on the path from a node to other computing power nodes.

The calculation formula for pheromone concentration includes:

Among them, δ _ij (t) is the pheromone concentration on the path from routing node i to routing node j corresponding to the request node; is the volatilization rate of pheromone, and/> v is the preset initial value of pheromone concentration; RSR _j is the rank sum ratio of computing node j; t _ij is the data transmission delay from routing node i to routing node j corresponding to the request node; A _cost is the preset computing power node charging method; m is an intermediate node that meets the preset requirements; λ ₁ and λ ₂ are both preset coefficients;/> Indicates the increase in pheromone concentration when a computing power request passes through the path from routing node i to routing node j. The value of the increase in pheromone concentration is affected by the service capability RSR _j of the computing power node corresponding to routing node j and the data transmission delay t The impact of _ij and node billing method A _cost .

As time goes by, the pheromone concentration on each path is constantly updated. In one implementation, the optimal target computing power node for providing services can be determined by selecting the path with the highest pheromone concentration.

S340: Determine the probability that other computing nodes serve as target computing nodes based on the data transmission delay between the requesting node and other computing nodes in the computing network and the pheromone concentration on the path from the requesting node to other computing nodes. value, and determine the computing power node corresponding to the largest probability value as the target computing power node.

Through S220, the shortest distance and the optimal path from routing node i to each routing node can be obtained. However, in actual business, it is not the optimal strategy to select the target computing node with the shortest distance to provide services to users every time. It is necessary to comprehensively consider the service capabilities of the computing power nodes. To this end, this manual uses an improved ant colony algorithm to select the optimal computing power nodes.

For example, based on the data transmission delay between the requesting node and other computing power nodes in the computing power network, and the pheromone concentration on the path from the requesting node to other computing power nodes, determine the probability that other computing power nodes are the target computing power nodes. value; determine the computing power node corresponding to the largest probability value as the target computing power node.

For example, when using the improved ant colony algorithm to select the target computing power node, the probability that a user at routing node i selects the computing power node at routing node j as the target computing power node at time t can be calculated by the following formula:

Among them, p _ij (t) is the probability value of the computing power node corresponding to routing node j as the target computing power node of the computing power node corresponding to routing node i; γ _ij (t) is the rank from requesting node i to computing power node j ; δ _ij (t) is the pheromone concentration on the path from routing node i corresponding to the request node to routing node j; k is the number of the routing node corresponding to the computing power node in the summation operator; α is the pheromone heuristic formula factor; β is the expectation heuristic factor.

To sum up, the computing power node selection method provided by the embodiment of this specification obtains multiple index data of each computing power node in the computing power network to determine the rank sum ratio set of each computing power node, where the computing power node Multiple index data are used to characterize the service capabilities of computing power nodes; obtain the data transmission delay between the requesting node and other computing power nodes in the computing power network; the requesting node is the node that issues the computing request among each computing power node; according to the pre-set Based on the set initial value of pheromone concentration, the set of rank sum ratios, and the data transmission delay between the requesting node and other computing nodes in the computing network, determine the pheromone concentration on the path from the requesting node to other computing nodes; according to The data transmission delay between the requesting node and other computing nodes in the computing network, and the pheromone concentration on the path from the requesting node to other computing nodes, determine the target computing node that will provide computing services to the requesting node. It can obtain multiple indicator data of each computing power node in the computing power network, determine the rank sum ratio set of computing power nodes, and obtain the data transmission delay between the requesting node and other computing power nodes. Based on these data and preset The initial value of the pheromone concentration can determine the best path from the requesting node to other computing nodes, and find the target computing node that is most suitable for providing computing services. This optimization process can be used under the limited computing resources and bandwidth of the computing system. Under the conditions, the user response delay is reduced, the efficiency and performance of the computing network are improved, and the requesting node receives high-quality computing services.

Figure 4 is a schematic structural diagram of a computing power node selection device 400 provided by an embodiment of this specification, including:

The data acquisition unit 410 acquires multiple index data of each computing power node in the computing power network to determine the rank sum ratio set of each computing power node, where the multiple index data of the computing power node is used to characterize the computing power node. Service capabilities of power nodes;

The delay obtaining unit 420 obtains the data transmission delay between the requesting node and other computing power nodes in the computing power network; the requesting node is the node that issues the computing request among the computing power nodes;

The processing unit 430 determines the requesting node based on the preset initial value of the pheromone concentration, the rank sum ratio set, and the data transmission delay between the requesting node and other computing power nodes in the computing power network. Pheromone concentration on the path to other computing power nodes;

Determining unit 440 determines, based on the data transmission delay between the requesting node and other computing power nodes in the computing power network, and the pheromone concentration on the path from the requesting node to other computing power nodes, that the computing power node is about to be processed. The target computing power node that requests the node to provide computing services.

Optionally, in one implementation, the determining unit 440 is used to:

According to the data transmission delay between the requesting node and other computing power nodes in the computing power network, and the pheromone concentration on the path from the requesting node to other computing power nodes, the other computing power nodes are determined as The probability value of the target computing power node;

The computing power node corresponding to the largest probability value is determined as the target computing power node.

Optionally, in one implementation, the data acquisition unit 410 is used to:

Obtain multiple indicator data of each computing power node; the multiple indicator data include CPU utilization, memory utilization, disk utilization and number of connections of each computing power node;

According to the original index data matrix composed of multiple index data of each computing power node, determine a rank data matrix composed of the ranks of each computing power node;

The rank sum ratio of each computing power node is determined according to the rank data matrix, and the rank sum ratio set is obtained.

Optionally, in one implementation, the delay acquisition unit 420 is used to:

Obtain the link bandwidth and transmission rate between routing nodes on the path between the requesting node and other computing power nodes in the computing power network;

Determine the data transmission delay between the routing nodes according to the ratio of the link bandwidth and the transmission rate between the routing nodes;

According to the data transmission delay between the routing nodes, the data transmission delay between the requesting node and other computing power nodes in the computing power network is determined.

Optionally, in one implementation, the delay acquisition unit 420 is used to:

Determine the distance set from the routing node corresponding to the requesting node to other routing nodes in the computing power network based on the delay matrix formed by the data transmission delay between the routing nodes;

Select an intermediate node that meets the preset requirements from the distance set;

The distance update step is performed based on the intermediate node; the distance update step includes: calculating the first distance from the routing node corresponding to the request node to the corresponding routing node of other computing power nodes in the computing power network through the intermediate node. Set, compare the first distance set with the distance in the distance set, and if the distance between the corresponding routing nodes in the first distance set is less than the distance in the distance set, use The distance values in the first distance set update the distance values in the distance set;

Select the next intermediate node that meets the preset requirements from the distance set, and repeat the distance update step until the distance values in the distance set meet the preset convergence conditions, and a target distance set is obtained.

Optionally, in one implementation, the processing unit 430 is used to:

According to the preset computing power node charging method, the preset pheromone concentration initial value, the rank sum ratio set, and the data between the requesting node and other computing power nodes in the computing power network The transmission delay determines the pheromone concentration on the path from the requesting node to other computing power nodes.

Alternatively, in one embodiment, the calculation formula of the pheromone concentration includes:

Among them, δ _ij (t) is the pheromone concentration on the path from routing node i corresponding to the request node to routing node j; is the volatilization rate of pheromone, and/> v is the preset initial value of pheromone concentration; RSR _j is the rank sum ratio of computing power node j; t _ij is the data transmission delay from routing node i to routing node j corresponding to the request node; A _cost is the preset computing power node charging method; m is an intermediate node that meets the preset requirements; λ ₁ and λ ₂ are both preset coefficients; Indicates the increase in pheromone concentration when a computing power request passes through the path from routing node i to routing node j. The value of the increase in pheromone concentration is affected by the service capability RSR j of the computing power node corresponding to routing node _j. , the impact of data transmission delay t _ij and node charging method A _cost .

Alternatively, in one implementation, the calculation formula for the probability value of the other computing power nodes as the target computing power node includes:

Among them, p _ij (t) is the probability value of the computing power node corresponding to the routing node j as the target computing power node of the computing power node corresponding to the routing node i; γ _ij (t) is the request node i to the computing power node j _The rank of ; α is the pheromone heuristic factor; β is the expectation heuristic factor.

The computing power node selection device 400 can implement the methods of the method embodiments in Figures 2 to 3. For details, reference can be made to the computing power node selection methods of the embodiments shown in Figures 2 to 3, which will not be described again.

FIG. 5 is a schematic structural diagram of an electronic device according to an embodiment of this specification. Please refer to Figure 5. At the hardware level, the electronic device includes a processor and optionally an internal bus, a network interface, and a memory. The memory may include memory, such as high-speed random access memory (Random-Access Memory, RAM), or may also include non-volatile memory (non-volatile memory), such as at least one disk memory. Of course, the electronic equipment may also include other hardware required by the business.

The processor, network interface and memory can be connected to each other through an internal bus, which can be an ISA (Industry Standard Architecture) bus, a PCI (Peripheral Component Interconnect, a peripheral component interconnect standard) bus or an EISA (Extended Industry StandardArchitecture, extended industry standard architecture) bus, etc. The bus can be divided into address bus, data bus, control bus, etc. For ease of presentation, only one bidirectional arrow is used in Figure 5, but it does not mean that there is only one bus or one type of bus.

Memory, used to store programs. Specifically, a program may include program code including computer operating instructions. Memory may include internal memory and non-volatile memory and provides instructions and data to the processor.

The processor reads the corresponding computer program from the non-volatile memory into the memory and then runs it, forming a computing power node selection device at the logical level. The processor executes the program stored in the memory and is specifically used to perform the following operations:

Obtain multiple index data of each computing power node in the computing power network to determine the rank sum ratio set of each computing power node, where the multiple index data of the computing power node is used to characterize the service capability of the computing power node ;

Obtain the data transmission delay between the requesting node and other computing power nodes in the computing power network; the requesting node is the node that issues the computing request among the computing power nodes;

According to the preset initial value of pheromone concentration, the rank sum ratio set and the data transmission delay between the requesting node and other computing power nodes in the computing power network, determine the distance between the requesting node and other computing power nodes. Pheromone concentration along the path of the node;

According to the data transmission delay between the requesting node and other computing power nodes in the computing power network, and the pheromone concentration on the path from the requesting node to other computing power nodes, it is determined that the requesting node will be provided with The target computing power node of the computing service.

It can obtain multiple indicator data of each computing power node in the computing power network, determine the rank sum ratio set of computing power nodes, and obtain the data transmission delay between the requesting node and other computing power nodes. Based on these data and preset The initial value of the pheromone concentration can determine the best path from the requesting node to other computing nodes, and find the target computing node that is most suitable for providing computing services. This optimization process can be used under the limited computing resources and bandwidth of the computing system. Under the conditions, the user response delay is reduced, the efficiency and performance of the computing network are improved, and the requesting node receives high-quality computing services.

The method performed by the computing power node selection device disclosed in the embodiment shown in FIG. 4 of this specification can be applied to a processor, or implemented by the processor. The processor may be an integrated circuit chip that has signal processing capabilities. During the implementation process, each step of the above method can be completed by instructions in the form of hardware integrated logic circuits or software in the processor. The above-mentioned processor can be a general-purpose processor, including a central processing unit (CPU), a network processor (Network Processor, NP), etc.; it can also be a digital signal processor (Digital SignalProcessor, DSP), an application-specific integrated circuit (Application Specific Integrated Circuit, ASIC), Field-Programmable Gate Array (FPGA) or other programmable logic devices, discrete gate or transistor logic devices, and discrete hardware components. Each method, step and logical block diagram disclosed in the embodiments of this specification can be implemented or executed. A general-purpose processor may be a microprocessor or the processor may be any conventional processor, etc. The steps of the methods disclosed in conjunction with the embodiments of this specification can be directly implemented by a hardware decoding processor, or executed by a combination of hardware and software modules in the decoding processor. The software module can be located in random access memory, flash memory, read-only memory, programmable read-only memory or electrically erasable programmable memory, registers and other mature storage media in this field. The storage medium is located in the memory, and the processor reads the information in the memory and completes the steps of the above method in combination with its hardware.

The electronic device can also perform the method in Figure 2 or Figure 3, and implement the functions of the computing power node selection device in the embodiment shown in Figure 4. The embodiments of this specification will not be described again here.

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

Embodiments of this specification also provide a computer-readable storage medium that stores one or more programs. The one or more programs include instructions. The instructions are used by a portable electronic device including multiple application programs. When executed, the portable electronic device can be caused to execute the method of the embodiment shown in Figure 3, and is specifically used to perform the following operations:

Obtain multiple index data of each computing power node in the computing power network to determine the rank sum ratio set of each computing power node, where the multiple index data of the computing power node is used to characterize the service capability of the computing power node ;

Obtain the data transmission delay between the requesting node and other computing power nodes in the computing power network; the requesting node is the node that issues the computing request among the computing power nodes;

According to the preset initial value of pheromone concentration, the rank sum ratio set and the data transmission delay between the requesting node and other computing power nodes in the computing power network, determine the distance between the requesting node and other computing power nodes. Pheromone concentration along the path of the node;

According to the data transmission delay between the requesting node and other computing power nodes in the computing power network, and the pheromone concentration on the path from the requesting node to other computing power nodes, it is determined that the requesting node will be provided with The target computing power node of the computing service.

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

The foregoing describes specific embodiments of this specification. Other embodiments are within the scope of the appended claims. In some cases, the actions or steps recited in the claims can be performed in a different order than in the embodiments and still achieve desired results. Additionally, the processes depicted in the figures do not necessarily require the specific order shown, or sequential order, to achieve desirable results. Multitasking and parallel processing are also possible or may be advantageous in certain implementations.

In short, the above descriptions are only preferred embodiments of this specification and are not intended to limit the scope of protection of this specification. Any modifications, equivalent replacements, improvements, etc. made within the spirit and principles of this manual shall be included in the protection scope of this manual.

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

Computer-readable media includes both persistent and non-volatile, removable and non-removable media that can be implemented by any method or technology for storage of information. Information may 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), and read-only memory. (ROM), electrically erasable programmable read-only memory (EEPROM), flash memory or other memory technology, compact disc read-only memory (CD-ROM), digital versatile disc (DVD) or other optical storage, Magnetic tape cassettes, tape magnetic disk storage or other magnetic storage devices or any other non-transmission medium can be used to store information that can be accessed by a computing device. As defined in this article, computer-readable media does not include transitory media, such as modulated data signals and carrier waves.

It should also be noted that the terms "comprises," "comprises" or any other variation thereof are intended to cover a non-exclusive inclusion, such that a process, method, article, or apparatus that includes a list of elements not only includes those elements, but also includes Other elements are not expressly listed or are inherent to the process, method, article or equipment. Without further limitation, an element defined by the statement "comprises a..." does not exclude the presence of additional identical elements in a process, method, article, or device that includes the stated element.

Each embodiment in this specification is described in a progressive manner. The same and similar parts between the various embodiments can be referred to each other. Each embodiment focuses on its differences from other embodiments. In particular, for the system embodiment, since it is basically similar to the method embodiment, the description is relatively simple. For relevant details, please refer to the partial description of the method embodiment.

Claims (11)

1. A computing power node selection method, characterized in that the method includes: Obtain multiple index data of each computing power node in the computing power network to determine the rank sum ratio set of each computing power node, where the multiple index data of the computing power node is used to characterize the service capability of the computing power node ; Obtain the data transmission delay between the requesting node and other computing power nodes in the computing power network; the requesting node is the node that issues the computing request among the computing power nodes; According to the preset initial value of pheromone concentration, the rank sum ratio set and the data transmission delay between the requesting node and other computing power nodes in the computing power network, determine the distance between the requesting node and other computing power nodes. Pheromone concentration along the path of the node; According to the data transmission delay between the requesting node and other computing power nodes in the computing power network, and the pheromone concentration on the path from the requesting node to other computing power nodes, it is determined that the requesting node will be provided with The target computing power node of the computing service. 2. The method according to claim 1, characterized in that, according to the data transmission delay between the requesting node and other computing power nodes in the computing power network, and the data transmission delay from the requesting node to other computing power nodes. The pheromone concentration on the path determines the target computing node that will provide computing services to the requesting node, including: According to the data transmission delay between the requesting node and other computing power nodes in the computing power network, and the pheromone concentration on the path from the requesting node to other computing power nodes, the other computing power nodes are determined as The probability value of the target computing power node; The computing power node corresponding to the largest probability value is determined as the target computing power node. 3. The method according to claim 1, characterized in that, obtaining multiple index data of each computing power node in the computing power network to determine the rank sum ratio set of each computing power node, including: Obtain multiple indicator data of each computing power node; the multiple indicator data include CPU utilization, memory utilization, disk utilization and number of connections of each computing power node; According to the original index data matrix composed of multiple index data of each computing power node, determine a rank data matrix composed of the ranks of each computing power node; The rank sum ratio of each computing power node is determined according to the rank data matrix, and the rank sum ratio set is obtained. 4. The method of claim 1, wherein obtaining the data transmission delay between the requesting node and other computing power nodes in the computing power network includes: Obtain the link bandwidth and transmission rate between routing nodes on the path between the requesting node and other computing power nodes in the computing power network; Determine the data transmission delay between the routing nodes according to the ratio of the link bandwidth and the transmission rate between the routing nodes; According to the data transmission delay between the routing nodes, the data transmission delay between the requesting node and other computing power nodes in the computing power network is determined. 5. The method of claim 4, wherein the data transmission time between the requesting node and other computing power nodes in the computing power network is determined based on the data transmission delay between the routing nodes. After postponement, include: Determine the distance set from the routing node corresponding to the requesting node to other routing nodes in the computing power network based on the delay matrix formed by the data transmission delay between the routing nodes; Select an intermediate node that meets the preset requirements from the distance set; The distance update step is performed based on the intermediate node; the distance update step includes: calculating the first distance from the routing node corresponding to the request node to the corresponding routing node of other computing power nodes in the computing power network through the intermediate node. Set, compare the first distance set with the distance in the distance set, and if the distance between the corresponding routing nodes in the first distance set is less than the distance in the distance set, use The distance values in the first distance set update the distance values in the distance set; Select the next intermediate node that meets the preset requirements from the distance set, and repeat the distance update step until the distance values in the distance set meet the preset convergence conditions, and a target distance set is obtained. 6. The method according to claim 1, characterized in that, according to the preset initial value of pheromone concentration, the rank sum ratio set and the relationship between the requesting node and other computing power nodes in the computing power network. The data transmission delay is determined by determining the pheromone concentration on the path from the requesting node to other computing nodes, including: According to the preset computing power node charging method, the preset pheromone concentration initial value, the rank sum ratio set, and the data between the requesting node and other computing power nodes in the computing power network The transmission delay determines the pheromone concentration on the path from the requesting node to other computing power nodes. 7. The method of claim 6, wherein the calculation formula of the pheromone concentration includes: Among them, δ _ij (t) is the pheromone concentration on the path from routing node i corresponding to the request node to routing node j; is the volatilization rate of pheromone, and/> v is the preset initial value of pheromone concentration; RSR _j is the rank sum ratio of computing power node j; t _ij is the data transmission delay from routing node i to routing node j corresponding to the request node; A _cost is the preset computing power node charging method; m is an intermediate node that meets the preset requirements; λ ₁ and λ ₂ are both preset coefficients;/> Indicates the increase in pheromone concentration when a computing power request passes through the path from routing node i to routing node j. The value of the increase in pheromone concentration is affected by the service capability RSR j of the computing power node corresponding to routing node _j. , the impact of data transmission delay t _ij and node charging method A _cost . 8. The method of claim 2, wherein the calculation formula for the probability value of the other computing power nodes as the target computing power node includes: Among them, p _ij (t) is the probability value of the computing power node corresponding to the routing node j as the target computing power node of the computing power node corresponding to the routing node i; γ _ij (t) is the request node i to the computing power node j _The rank of ; α is the pheromone heuristic factor; β is the expectation heuristic factor. 9. A computing power node selection device, characterized by including: The data acquisition unit acquires multiple index data of each computing power node in the computing power network to determine the rank sum ratio set of each computing power node, wherein the multiple index data of the computing power node is used to characterize the computing power. The service capability of the node; A delay acquisition unit obtains the data transmission delay between the requesting node and other computing nodes in the computing power network; the requesting node is the node that issues the computing request among the computing power nodes; The processing unit determines the arrival of the requesting node based on the preset initial value of the pheromone concentration, the rank sum ratio set, and the data transmission delay between the requesting node and other computing power nodes in the computing power network. Pheromone concentration on the path of other computing power nodes; The determining unit determines the upcoming data transmission time between the requesting node and other computing power nodes in the computing power network and the pheromone concentration on the path from the requesting node to other computing power nodes. The target computing power node that requests the node to provide computing services. 10. An electronic device, characterized in that it includes: processor; and Memory arranged to store computer-executable instructions that, when executed, cause the processor to: Obtain multiple index data of each computing power node in the computing power network to determine the rank sum ratio set of each computing power node, where the multiple index data of the computing power node is used to characterize the service capability of the computing power node ; Obtain the data transmission delay between the requesting node and other computing power nodes in the computing power network; the requesting node is the node that issues the computing request among the computing power nodes; According to the preset initial value of pheromone concentration, the rank sum ratio set and the data transmission delay between the requesting node and other computing power nodes in the computing power network, determine the distance between the requesting node and other computing power nodes. Pheromone concentration along the path of the node; According to the data transmission delay between the requesting node and other computing power nodes in the computing power network, and the pheromone concentration on the path from the requesting node to other computing power nodes, it is determined that the requesting node will be provided with The target computing power node of the computing service. 11. A computer-readable storage medium, characterized in that the computer-readable storage medium stores one or more programs, and the one or more programs, when executed by an electronic device including a plurality of application programs, cause the The electronic device described above performs the following operations: Obtain multiple index data of each computing power node in the computing power network to determine the rank sum ratio set of each computing power node, where the multiple index data of the computing power node is used to characterize the service capability of the computing power node ; Obtain the data transmission delay between the requesting node and other computing power nodes in the computing power network; the requesting node is the node that issues the computing request among the computing power nodes; According to the preset initial value of pheromone concentration, the rank sum ratio set and the data transmission delay between the requesting node and other computing power nodes in the computing power network, determine the distance between the requesting node and other computing power nodes. Pheromone concentration along the path of the node; According to the data transmission delay between the requesting node and other computing power nodes in the computing power network, and the pheromone concentration on the path from the requesting node to other computing power nodes, it is determined that the requesting node will be provided with The target computing power node of the computing service.