Computing power node selection method, device and equipment, medium and program product

Abstract

The invention discloses a computing power node selection method, device and equipment, a medium and a program product, computing power detection is carried out according to each computing power type of a to-be-measured force node, and it is ensured that acquired computing power detection information can accurately reflect actual computing power of each computing power type of the to-be-measured force node, so that the computing power of each computing power type of the to-be-measured force node can be accurately reflected when the computing power node is selected. It can be ensured that the actual computing power of the selected target computing power node can effectively meet the computing power requirement of a current user, and the difference between the computing power actually obtained by the user and computing power detection information is effectively reduced.

Description

Technical Field

[0001] This invention relates to the field of computing network technology, and in particular to a method, apparatus, electronic device, computer-readable storage medium, and computer program product for selecting computing nodes. Background Technology

[0002] A computing power network is a new type of information infrastructure that allocates and flexibly schedules computing, storage, and network resources on demand among the cloud, network, and edge based on business needs. Due to differences in construction time, hardware selection, and software selection, the actual computing capabilities of computing power nodes in a computing power network vary significantly. Therefore, it is necessary to provide a method for selecting computing power nodes to measure the actual computing power of each node and meet the computing power needs of different users.

[0003] Existing methods for selecting computing nodes typically calculate absolute computing power based on parameters such as the CPU (Central Processing Unit) clock speed, cache, and bandwidth of the computing node. However, computing nodes can be oversold, which can lead to a difference between the actual computing power obtained by the user and the calculated absolute computing power, resulting in poor accuracy in selecting computing nodes. Summary of the Invention

[0004] This invention provides a method, apparatus, device, medium, and program product for selecting computing nodes, in order to solve the technical problem of poor accuracy in selecting computing nodes in the prior art.

[0005] To address the aforementioned technical problems, a first aspect of this invention provides a method for selecting computing nodes, comprising: According to the preset computing power types, computing power detection is performed on any computing power node to be tested that meets the current user's computing power requirements and has not undergone computing power detection, so as to obtain computing power detection information for each computing power type corresponding to the computing power node to be tested. Based on the current user's computing power requirements and the currently acquired computing power detection information, determine the current computing power detection execution probability; Repeat the aforementioned steps until the current computing power detection execution probability meets the preset computing power detection end condition, and obtain at least one candidate computing power node; wherein, the candidate computing power node is the computing power node to be tested that has undergone computing power detection; Based on the computing power detection information of each candidate computing power node, the target computing power node corresponding to the current user's computing power requirement is determined.

[0006] As a preferred embodiment, the step of performing computing power detection on any computing power node that meets the current user's computing power requirements and has not undergone computing power detection, according to preset computing power types, to obtain computing power detection information for each of the computing power types corresponding to the computing power node under test, specifically: Based on the preset image addresses of computing power tasks corresponding to each computing power type, any one of the computing power nodes under test is instructed to run the image computing power tasks corresponding to each computing power task image address, to obtain the computing power value of each computing power type corresponding to the computing power node under test, and to use the computing power value of each computing power type as the computing power detection information of each computing power type corresponding to the computing power node under test.

[0007] As a preferred embodiment, determining the current computing power detection execution probability based on the current user's computing power demand and the currently acquired computing power detection information specifically includes: Based on the preset maximum computing power usage cost corresponding to each of the computing power types and the computing power value, the unit computing power usage cost corresponding to each of the computing power types of the computing power node to be tested is determined; Based on the current user computing power demand and the unit computing power usage cost of each computing power type corresponding to each of the current computing power nodes to be tested, determine the current required computing power usage cost of each of the current computing power nodes to be tested; Based on the current computing power usage cost of each of the aforementioned requirements, the current probability of executing the computing power probe is determined.

[0008] As a preferred embodiment, determining the current computing power detection execution probability based on the current computing power usage cost of each of the aforementioned demands specifically includes: Based on the current computing power usage cost of each of the aforementioned demand computing power nodes, determine the average computing power usage cost of each of the computing power nodes to be tested before the current computing power detection. Based on the required computing power usage cost and the average required computing power usage cost of the computing power node corresponding to the current computing power detection, determine the current rate of change of required computing power usage cost; Based on the current rate of change and rate of change index of computing power usage cost, the current probability of executing computing power detection is determined; wherein, the current rate of change index is equal to the current number of computing power detection executions.

[0009] As a preferred embodiment, the specific condition for ending the computing power detection is: the execution probability of the computing power detection is less than the current random number; wherein, the random number is obtained by randomly selecting from the interval of 0 to 1 after the current computing power detection.

[0010] As a preferred embodiment, determining the target computing power node corresponding to the current user's computing power requirement based on the computing power detection information of each of the candidate computing power nodes specifically includes: Based on the computing power detection information of each candidate computing power node, the required computing power usage cost of each candidate computing power node is determined. The candidate computing power node with the lowest cost of using the required computing power is selected as the target computing power node.

[0011] A second aspect of the present invention provides a computing node selection device, comprising: The computing power detection module is used to perform computing power detection on any computing power node to be tested that meets the current user's computing power requirements and has not undergone computing power detection, according to preset computing power types, and to obtain computing power detection information for each computing power type corresponding to the computing power node to be tested. The computing power detection execution probability determination module is used to determine the current computing power detection execution probability based on the current user computing power demand and the currently acquired computing power detection information. The candidate computing power node determination module is used to repeat the aforementioned steps until the current computing power detection execution probability meets the preset computing power detection end condition, thereby obtaining at least one candidate computing power node; wherein, the candidate computing power node is the computing power node to be tested that has undergone computing power detection. The target computing power node determination module is used to determine the target computing power node corresponding to the current user's computing power requirement based on the computing power detection information of each of the candidate computing power nodes.

[0012] A third aspect of the present invention provides an electronic device, including a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor executes the computer program to implement the computing node selection method described in any of the first aspects.

[0013] A fourth aspect of the present invention provides a computer-readable storage medium comprising a stored computer program, wherein, when the computer program is executed, it controls the device where the computer-readable storage medium is located to perform the computing node selection method described in any of the first aspects.

[0014] A fifth aspect of the present invention provides a computer program product, including a computer program / instructions, wherein when the computer program / instructions are executed by a processor, the steps of the computing node selection method described in any of the first aspects are implemented.

[0015] Compared with the prior art, the beneficial effect of the embodiments of the present invention is that by performing computing power detection according to each computing power type of the computing power node to be tested, it ensures that the obtained computing power detection information can accurately reflect the actual computing power of each computing power type of the computing power node to be tested. Thus, when selecting computing power nodes, it can ensure that the actual computing power of the selected target computing power node can effectively meet the current computing power needs of the user, and effectively reduce the difference between the actual computing power obtained by the user and the computing power detection information. Attached Figure Description

[0016] Figure 1 This is a flowchart illustrating the computing node selection method in an embodiment of the present invention; Figure 2 This is a schematic diagram of the computing node selection device in an embodiment of the present invention; Figure 3 This is a schematic diagram of the structure of the electronic device in an embodiment of the present invention. Detailed Implementation

[0017] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0018] Please see Figure 1 The first aspect of this invention provides a method for selecting computing nodes, comprising the following steps S1 to S4: Step S1: According to the preset computing power types, perform computing power detection on any computing power node to be tested that meets the current user's computing power requirements and has not undergone computing power detection, and obtain computing power detection information for each computing power type corresponding to the computing power node to be tested. Step S2: Based on the current user's computing power requirements and the currently acquired computing power detection information, determine the current computing power detection execution probability; Step S3: Repeat the above steps until the current computing power detection execution probability meets the preset computing power detection end condition, and obtain at least one candidate computing power node; wherein, the candidate computing power node is the computing power node to be tested that has undergone computing power detection. Step S4: Based on the computing power detection information of each candidate computing power node, determine the target computing power node corresponding to the current user's computing power requirement.

[0019] Specifically, this embodiment first needs to determine the current user's computing power requirements, and then identify multiple computing power nodes that meet those requirements. To accurately obtain the actual computing power of these nodes and avoid a large discrepancy between the absolute computing power calculated based on static hardware parameters and the actual computing power obtained by the user, this embodiment randomly selects one computing power node to be tested from these nodes and performs computing power detection according to various computing power types. This refines the granularity of computing power detection for the node under test, accurately obtaining the computing power detection information corresponding to each computing power type. It is worth noting that the computing power types in this embodiment include, but are not limited to: integer operations, single-precision floating-point operations, double-precision floating-point operations, and sequential reads per second.

[0020] Furthermore, in order to simultaneously consider the efficiency of computing power detection, this embodiment determines the current computing power detection execution probability based on the current user's computing power demand and the currently acquired computing power detection information. This computing power detection execution probability is used to decide whether to continue to execute the next computing power detection, ensuring that the target computing power node corresponding to the current user's computing power demand is selected with as few computing power detection attempts as possible.

[0021] Furthermore, in this embodiment, computing power detection is repeatedly performed on the computing power nodes that meet the current user's computing power requirements but have not undergone computing power detection. The execution probability of computing power detection is continuously calculated after each detection. If the current execution probability of computing power detection does not meet the preset computing power detection termination condition, computing power detection is continued on the computing power nodes that meet the current user's computing power requirements but have not undergone computing power detection. If the current execution probability of computing power detection meets the preset computing power detection termination condition, it indicates that the probability of detecting a better computing power node is low. Considering the efficiency of computing power detection, computing power detection is stopped. At this time, at least one candidate computing power node is obtained. Then, based on the computing power detection information of these candidate computing power nodes, the target computing power node corresponding to the current user's computing power requirements is selected.

[0022] The computing power node selection method provided in this invention detects computing power according to each computing power type of the computing power node to be tested, ensuring that the obtained computing power detection information can accurately reflect the actual computing power of each computing power type of the computing power node to be tested. Thus, when selecting computing power nodes, it can ensure that the actual computing power of the selected target computing power node can effectively meet the current user's computing power needs, effectively reducing the difference between the actual computing power obtained by the user and the computing power detection information.

[0023] As a preferred embodiment, the step of performing computing power detection on any computing power node that meets the current user's computing power requirements and has not undergone computing power detection, according to preset computing power types, to obtain computing power detection information for each of the computing power types corresponding to the computing power node under test, specifically: Based on the preset image addresses of computing power tasks corresponding to each computing power type, any one of the computing power nodes under test is instructed to run the image computing power tasks corresponding to each computing power task image address, to obtain the computing power value of each computing power type corresponding to the computing power node under test, and to use the computing power value of each computing power type as the computing power detection information of each computing power type corresponding to the computing power node under test.

[0024] Specifically, in order to obtain the actual computing power of the computing power node under test for different computing power types, this embodiment first constructs the mirror computing power tasks corresponding to each computing power type and generates the mirror addresses of each computing power task. For example, the mirror addresses of computing power tasks for different computing power types can be shown in Table 1 below.

[0025] Table 1. Correspondence between computing power type and computing power task mirror address

[0026] The computing power types listed in Table 1 above are only partial examples. This embodiment does not specifically limit the computing power types to be detected, and the computing power task mirror addresses can also be set for different computing power types. Based on the computing power task mirror addresses for each computing power type, by instructing the computing power node under test to run the mirror computing power tasks corresponding to each computing power task mirror address, the current actual computing power of the computing power node under test for each computing power type can be accurately obtained, ensuring that the difference between the actual computing power obtained by the user and the computing power detection results is small after subsequent computing power node selection. Furthermore, the obtained computing power values ​​for each computing power type corresponding to the computing power node under test constitute a computing power vector: ,in, Indicates the use of computing power tasks The detected computing power value for the corresponding computing power type, where k is the number of computing power tasks, which is also equal to the number of computing power types.

[0027] As a preferred embodiment, determining the current computing power detection execution probability based on the current user's computing power demand and the currently acquired computing power detection information specifically includes: Based on the preset maximum computing power usage cost corresponding to each of the computing power types and the computing power value, the unit computing power usage cost corresponding to each of the computing power types of the computing power node to be tested is determined; Based on the current user computing power demand and the unit computing power usage cost of each computing power type corresponding to each of the current computing power nodes to be tested, determine the current required computing power usage cost of each of the current computing power nodes to be tested; Based on the current computing power usage cost of each of the aforementioned requirements, the current probability of executing the computing power probe is determined.

[0028] Specifically, the computing power value corresponding to each computing power type is denoted as: The maximum computing power usage cost corresponding to each computing power type is denoted as... The unit computing power cost for each computing power type corresponding to the computing power node to be tested can be denoted as: It is worth noting that the maximum computing power usage cost corresponding to each computing power type of each computing power node to be tested is different, and the maximum computing power usage cost corresponding to each computing power type is preset. This embodiment does not make specific limitations here.

[0029] Furthermore, based on current user computing power requirements: ,in, Indicates for the first The computing power requirement values ​​for each computing power type are calculated, thus enabling the determination of the computing power usage cost for the current user to complete the required computing power using each tested computing power node. The computing power usage cost for using the i-th tested computing power node to complete the required computing power is denoted as . The details are as follows: ; Therefore, the required computing power usage cost for the current n computing power nodes to be tested can be obtained, denoted as follows: , where n is the number of computing power nodes to be tested after computing power detection.

[0030] Furthermore, in order to achieve a balance between optimal detection efficiency and optimal cost of computing power usage, this embodiment determines the current probability of performing computing power detection based on the current cost of each required computing power usage.

[0031] As a preferred embodiment, determining the current computing power detection execution probability based on the current computing power usage cost of each of the aforementioned demands specifically includes: Based on the current computing power usage cost of each of the aforementioned demand computing power nodes, determine the average computing power usage cost of each of the computing power nodes to be tested before the current computing power detection. Based on the required computing power usage cost and the average required computing power usage cost of the computing power node corresponding to the current computing power detection, determine the current rate of change of required computing power usage cost; Based on the current rate of change and rate of change index of computing power usage cost, the current probability of executing computing power detection is determined; wherein, the current rate of change index is equal to the current number of computing power detection executions.

[0032] Specifically, this embodiment calculates the average required computing power usage cost of each computing power node to be tested before the current computing power detection using the following expression. : ; in, This represents the number of computing power nodes that have undergone computing power detection prior to the current computing power detection.

[0033] Furthermore, by comparing the required computing power usage cost of the computing power node to be tested corresponding to the current computing power probe... and average computing power usage cost It can determine the rate of change in current demand for computing power usage costs: Understandably, a lower rate of change in the cost of demanded computing power indicates a lower probability of subsequently detecting computing power nodes with even better costs. To avoid excessive computing power probing, this embodiment uses the current rate of change in the cost of demanded computing power and its rate of change exponent, where the rate of change exponent is equal to the current number of computing power probing executions. After each computing power probing execution, the rate of change exponent is increased to reduce the probability of performing computing power probing after each execution. ,Right now .

[0034] As a preferred embodiment, the specific condition for ending the computing power detection is: the execution probability of the computing power detection is less than the current random number; wherein, the random number is obtained by randomly selecting from the interval of 0 to 1 after the current computing power detection.

[0035] Specifically, in this embodiment, after the current computing power detection is completed, a random number is randomly selected from the interval 0 to 1 according to a uniform distribution. It can be understood that as the number of computing power detections increases, the execution probability of the detection decreases continuously. Therefore, the probability that the selected random number is greater than the execution probability increases continuously, thus ensuring the efficiency of the computing power detection. When the execution probability of the computing power detection is less than the current random number, the detection stops, and the target computing power node is selected from the candidate computing power nodes that have already been detected. When the execution probability of the computing power detection is greater than or equal to the current random number, the next computing power detection task is executed.

[0036] As a preferred embodiment, determining the target computing power node corresponding to the current user's computing power requirement based on the computing power detection information of each of the candidate computing power nodes specifically includes: Based on the computing power detection information of each candidate computing power node, the required computing power usage cost of each candidate computing power node is determined. The candidate computing power node with the lowest cost of using the required computing power is selected as the target computing power node.

[0037] Understandably, during the selection of computing power nodes, the required computing power usage cost corresponding to the computing power detection information of each computing power node to be tested has been confirmed. Therefore, after obtaining at least one candidate computing power node, the corresponding required computing power usage cost can be directly determined based on its computing power detection information, and the candidate computing power node with the lowest required computing power usage cost can be selected as the current target computing power node. In this way, the computing power node with the optimal required computing power usage cost can be selected with as few computing power detections as possible, so as to keep the user's required computing power usage cost as low as possible while meeting the user's current computing power needs.

[0038] Please see Figure 2 A second aspect of the present invention provides a computing node selection device 100, comprising: The computing power detection module 11 is used to perform computing power detection on any computing power node to be tested that meets the current user's computing power requirements and has not undergone computing power detection, according to preset computing power types, and to obtain computing power detection information for each computing power type corresponding to the computing power node to be tested. The computing power detection execution probability determination module 12 is used to determine the current computing power detection execution probability based on the current user computing power demand and the currently acquired computing power detection information. The candidate computing power node determination module 13 is used to repeat the aforementioned steps until the current computing power detection execution probability meets the preset computing power detection end condition, thereby obtaining at least one candidate computing power node; wherein, the candidate computing power node is the computing power node to be tested that has undergone computing power detection. The target computing power node determination module 14 is used to determine the target computing power node corresponding to the current user's computing power demand based on the computing power detection information of each of the candidate computing power nodes.

[0039] As a preferred embodiment, the computing power detection module 11 is used to perform computing power detection on any computing power node to be tested that meets the current user's computing power requirements and has not undergone computing power detection, according to preset computing power types, to obtain computing power detection information for each computing power type corresponding to the computing power node to be tested, specifically: Based on the preset image addresses of computing power tasks corresponding to each computing power type, any one of the computing power nodes under test is instructed to run the image computing power tasks corresponding to each computing power task image address, to obtain the computing power value of each computing power type corresponding to the computing power node under test, and to use the computing power value of each computing power type as the computing power detection information of each computing power type corresponding to the computing power node under test.

[0040] As a preferred embodiment, the computing power detection execution probability determination module 12 is used to determine the current computing power detection execution probability based on the current user computing power demand and the currently acquired computing power detection information, specifically including: Based on the preset maximum computing power usage cost corresponding to each of the computing power types and the computing power value, the unit computing power usage cost corresponding to each of the computing power types of the computing power node to be tested is determined; Based on the current user computing power demand and the unit computing power usage cost of each computing power type corresponding to each of the current computing power nodes to be tested, determine the current required computing power usage cost of each of the current computing power nodes to be tested; Based on the current computing power usage cost of each of the aforementioned requirements, the current probability of executing the computing power probe is determined.

[0041] As a preferred embodiment, the computing power detection execution probability determination module 12 is used to determine the current computing power detection execution probability based on the current computing power usage cost of each of the aforementioned demands, specifically including: Based on the current computing power usage cost of each of the aforementioned demand computing power nodes, determine the average computing power usage cost of each of the computing power nodes to be tested before the current computing power detection. Based on the required computing power usage cost and the average required computing power usage cost of the computing power node corresponding to the current computing power detection, determine the current rate of change of required computing power usage cost; Based on the current rate of change and rate of change index of computing power usage cost, the current probability of executing computing power detection is determined; wherein, the current rate of change index is equal to the current number of computing power detection executions.

[0042] As a preferred embodiment, the specific condition for ending the computing power detection is: the execution probability of the computing power detection is less than the current random number; wherein, the random number is obtained by randomly selecting from the interval of 0 to 1 after the current computing power detection.

[0043] As a preferred embodiment, the target computing power node determination module 14 is used to determine the target computing power node corresponding to the current user's computing power demand based on the computing power detection information of each of the candidate computing power nodes, specifically including: Based on the computing power detection information of each candidate computing power node, the required computing power usage cost of each candidate computing power node is determined. The candidate computing power node with the lowest cost of using the required computing power is selected as the target computing power node.

[0044] The computing power node selection device provided in this embodiment of the invention performs computing power detection according to each computing power type of the computing power node to be tested, ensuring that the obtained computing power detection information can accurately reflect the actual computing power of each computing power type of the computing power node to be tested. Thus, when selecting computing power nodes, it can ensure that the actual computing power of the selected target computing power node can effectively meet the current computing power needs of the user, effectively reducing the difference between the actual computing power obtained by the user and the computing power detection information.

[0045] Please see Figure 3 The third aspect of the present invention provides an electronic device 200, including a memory 22, a processor 21, and a computer program stored in the memory 22 and executable on the processor 21. When the processor 21 executes the computer program, it implements the computing node selection method described in any embodiment of the first aspect.

[0046] For example, the computer program may be divided into one or more modules / units, which are stored in the memory 22 and executed by the processor 21 to complete the present invention. The one or more modules / units may be a series of computer program instruction segments capable of performing a specific function, which describe the execution process of the computer program in the electronic device 200.

[0047] The electronic device 200 may include, but is not limited to, a processor 21 and a memory 22. Those skilled in the art will understand that the schematic diagram is merely an example of the electronic device 200 and does not constitute a limitation on the electronic device 200. It may include more or fewer components than illustrated, or combine certain components, or different components. For example, the electronic device 200 may also include input / output devices, network access devices, buses, etc.

[0048] The processor 21 can be a central processing unit (CPU), or other general-purpose processors, digital signal processors (DSPs), application-specific integrated circuits (ASICs), field-programmable gate arrays (FPGAs), or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc. The general-purpose processor can be a microprocessor, or the processor 21 can be any conventional processor 21. The processor 21 is the control center of the electronic device 200, connecting various parts of the electronic device 200 via various interfaces and lines.

[0049] The memory 22 can be used to store the computer programs and / or modules. The processor 21 implements various functions of the electronic device 200 by running or executing the computer programs and / or modules stored in the memory 22 and calling the data stored in the memory 22. The memory 22 may mainly include a program storage area and a data storage area. The program storage area may store the operating system, at least one application program required for a function (such as sound playback function, image playback function, etc.), etc.; the data storage area may store data created according to the use of the mobile phone (such as audio data, phonebook, etc.). In addition, the memory 22 may include high-speed random access memory, and may also include non-volatile memory, such as hard disk, memory, plug-in hard disk, smart media card (SMC), secure digital (SD) card, flash card, at least one disk storage device, flash memory device, or other volatile solid-state storage device.

[0050] A fourth aspect of the present invention provides a computer-readable storage medium comprising a stored computer program, wherein, when the computer program is executed, it controls the device where the computer-readable storage medium is located to perform the computing node selection method described in any embodiment of the first aspect.

[0051] A fifth aspect of the present invention provides a computer program product, including a computer program / instructions, wherein when the computer program / instructions are executed by a processor, the steps of the computing node selection method described in any embodiment of the first aspect are implemented.

[0052] Wherein, if the modules / units integrated in the electronic device are implemented as software functional units and sold or used as independent products, they can be stored in a computer-readable storage medium. Based on this understanding, all or part of the processes in the methods of the above embodiments of the present invention can also be implemented by a computer program instructing related hardware. The computer program can be stored in a computer-readable storage medium, and when executed by a processor, it can implement the steps of the various method embodiments described above. The computer program includes computer program code, which can be in the form of source code, object code, executable files, or certain intermediate forms. The computer-readable medium can include: any entity or device capable of carrying the computer program code, recording media, USB flash drives, portable hard drives, magnetic disks, optical disks, computer memory, read-only memory (ROM), random access memory (RAM), electrical carrier signals, telecommunication signals, and software distribution media, etc.

[0053] The above description represents the preferred embodiments of the present invention. It should be noted that those skilled in the art can make various improvements and modifications without departing from the principles of the present invention, and these improvements and modifications are also considered to be within the scope of protection of the present invention.

Claims

1. A method for selecting computing nodes, characterized in that, include: According to the preset computing power types, computing power detection is performed on any computing power node to be tested that meets the current user's computing power requirements and has not undergone computing power detection, so as to obtain computing power detection information for each computing power type corresponding to the computing power node to be tested. Based on the current user's computing power requirements and the currently acquired computing power detection information, determine the current computing power detection execution probability; Repeat the aforementioned steps until the current computing power detection execution probability meets the preset computing power detection end condition, and obtain at least one candidate computing power node; wherein, the candidate computing power node is the computing power node to be tested that has undergone computing power detection; Based on the computing power detection information of each of the candidate computing power nodes, the target computing power node corresponding to the current user's computing power requirement is determined.

2. The computing node selection method as described in claim 1, characterized in that, The step involves performing computing power detection on any computing power node that meets the current user's computing power requirements but has not undergone computing power detection, according to preset computing power types, to obtain computing power detection information for each of the aforementioned computing power types corresponding to the computing power node under test. Specifically: Based on the preset image addresses of computing power tasks corresponding to each computing power type, any one of the computing power nodes under test is instructed to run the image computing power tasks corresponding to each computing power task image address, to obtain the computing power value of each computing power type corresponding to the computing power node under test, and to use the computing power value of each computing power type as the computing power detection information of each computing power type corresponding to the computing power node under test.

3. The computing node selection method as described in claim 2, characterized in that, The determination of the current computing power detection execution probability based on the current user's computing power demand and the currently acquired computing power detection information specifically includes: Based on the preset maximum computing power usage cost corresponding to each of the computing power types and the computing power value, the unit computing power usage cost corresponding to each of the computing power types of the computing power node to be tested is determined; Based on the current user computing power demand and the unit computing power usage cost of each computing power type corresponding to each of the current computing power nodes to be tested, determine the current required computing power usage cost of each of the current computing power nodes to be tested; Based on the current computing power usage cost of each of the aforementioned requirements, the current probability of executing the computing power probe is determined.

4. The computing node selection method as described in claim 3, characterized in that, The determination of the current computing power detection execution probability based on the current computing power usage cost of each of the aforementioned demands specifically includes: Based on the current computing power usage cost of each of the aforementioned demand computing power nodes, determine the average computing power usage cost of each of the computing power nodes to be tested before the current computing power detection. Based on the required computing power usage cost and the average required computing power usage cost of the computing power node corresponding to the current computing power detection, determine the current rate of change of required computing power usage cost; Based on the current rate of change and rate of change index of computing power usage cost, the current probability of executing computing power detection is determined; wherein, the current rate of change index is equal to the current number of computing power detection executions.

5. The computing node selection method as described in claim 1, characterized in that, The specific condition for ending the computing power detection is that the execution probability of the computing power detection is less than the current random number; wherein the random number is obtained by randomly selecting from the interval of 0 to 1 after the current computing power detection.

6. The computing node selection method as described in claim 3, characterized in that, The step of determining the target computing power node corresponding to the current user's computing power demand based on the computing power detection information of each of the candidate computing power nodes specifically includes: Based on the computing power detection information of each candidate computing power node, the required computing power usage cost of each candidate computing power node is determined. The candidate computing power node with the lowest cost of using the required computing power is selected as the target computing power node.

7. A computing node selection device, characterized in that, include: The computing power detection module is used to perform computing power detection on any computing power node to be tested that meets the current user's computing power requirements and has not undergone computing power detection, according to preset computing power types, and to obtain computing power detection information for each computing power type corresponding to the computing power node to be tested. The computing power detection execution probability determination module is used to determine the current computing power detection execution probability based on the current user computing power demand and the currently acquired computing power detection information. The candidate computing power node determination module is used to repeat the aforementioned steps until the current computing power detection execution probability meets the preset computing power detection end condition, thereby obtaining at least one candidate computing power node; wherein, the candidate computing power node is the computing power node to be tested that has undergone computing power detection. The target computing power node determination module is used to determine the target computing power node corresponding to the current user's computing power requirement based on the computing power detection information of each of the candidate computing power nodes.

8. An electronic device, characterized in that, It includes a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor executes the computer program to implement the computing node selection method according to any one of claims 1 to 6.

9. A computer-readable storage medium, characterized in that, The computer-readable storage medium includes a stored computer program, wherein, when the computer program is executed, it controls the device where the computer-readable storage medium is located to perform the computing node selection method according to any one of claims 1 to 6.

10. A computer program product, characterized in that, It includes a computer program / instruction that, when executed by a processor, implements the steps of the computing node selection method according to any one of claims 1 to 6.

Images