Computing power analysis method, computing power analysis apparatus, and storage medium

Abstract

The invention discloses a computing power analysis method, computing power analysis equipment and a storage medium, and relates to the technical field of data processing, and the method comprises the steps: obtaining computing power resource pool information corresponding to a computing power task; based on a first optimization target, analyzing the computing power resource pool information by using a simulated annealing algorithm to obtain a plurality of candidate scheduling schemes; constructing a flow control parameter of a genetic algorithm according to the plurality of candidate scheduling schemes, and analyzing the plurality of candidate scheduling schemes by using the genetic algorithm based on a second optimization target and the flow control parameter to obtain a target scheduling scheme; and executing computing power scheduling according to the target scheduling scheme. The invention aims to improve the accuracy of computing power scheduling so as to accurately meet the computing power demand of a user.

Description

Technical Field

[0001] This application relates to the technical field of data processing, and particularly to a computing power analysis method, a computing power analysis device, and a storage medium. Background Art

[0002] During the computing power leasing process, it is generally necessary to first construct a computing power resource cluster, and then select a scheduling scheme for the computing power resources that meet the computing task requirements according to a certain algorithm for computing power scheduling.

[0003] However, currently, in the process of analyzing the computing power scheduling scheme, it is generally optimized through a single fixed algorithm. Such a method is limited by the limitations of the selected algorithm, resulting in a problem of poor accuracy in the analyzed computing power scheduling scheme and being unable to meet the computing power requirements of users. Summary of the Invention

[0004] The main purpose of this application is to provide a computing power analysis method, a computing power analysis device, and a storage medium, aiming to improve the accuracy of computing power scheduling to accurately meet the computing power requirements of users.

[0005] To achieve the above object, this application proposes a computing power analysis method, and the method includes: Obtain the computing power resource pool information corresponding to the computing power task; Based on the first optimization goal, use the simulated annealing algorithm to analyze the computing power resource pool information to obtain multiple candidate scheduling schemes; Construct the process control parameters of the genetic algorithm according to the multiple candidate scheduling schemes, and based on the second optimization goal and the process control parameters, use the genetic algorithm to analyze the multiple candidate scheduling schemes to obtain the target scheduling scheme; Execute the computing power scheduling according to the target scheduling scheme.

[0006] In one embodiment, the step of constructing the process control parameters of the genetic algorithm according to the multiple candidate scheduling schemes includes: Determine the initial population in the genetic algorithm according to the multiple candidate scheduling schemes, and adjust the preset fitness function of the genetic algorithm corresponding to the second optimization goal according to the multiple candidate scheduling schemes to obtain the target fitness function; Wherein, the process control parameters include the initial population and the target fitness function.

[0007] In one embodiment, the first optimization goal includes minimizing the scheduling cost and maximizing the scheduling speed. The step of adjusting the preset fitness function of the genetic algorithm corresponding to the second optimization goal according to the multiple candidate scheduling schemes to obtain the target fitness function includes: Determine the minimum cost among the scheduling costs corresponding to all candidate scheduling schemes, and determine the maximum speed among the scheduling speeds corresponding to all candidate scheduling schemes; Determine the cost difference between the minimum cost and the preset minimum cost corresponding to the second optimization goal, and determine the speed difference between the maximum speed and the preset maximum speed corresponding to the second optimization goal; Adjust the preset fitness function according to the cost difference and the speed difference to obtain the target fitness function.

[0008] In one embodiment, the step of analyzing the computing power resource pool information by using the simulated annealing algorithm based on the first optimization goal to obtain multiple candidate scheduling schemes includes: Based on the first optimization goal, use the simulated annealing algorithm to determine the optimal solution among the first scheduling scheme and the second scheduling scheme corresponding to the computing power resource pool information; When the iteration end condition is not satisfied, use the current optimal solution as the new second scheduling scheme, determine the new scheduling scheme in the computing power resource pool information as the new first scheduling scheme, and return to execute the step of using the simulated annealing algorithm to determine the optimal solution among the first scheduling scheme and the second scheduling scheme corresponding to the computing power resource pool information based on the first optimization goal; When the iteration end condition is satisfied, determine the multiple candidate scheduling schemes among the preset number of multiple optimal solutions generated latest before the iteration end condition is satisfied.

[0009] In one embodiment, the step of determining the multiple candidate scheduling schemes among the preset number of multiple optimal solutions generated latest before the iteration end condition is satisfied includes: Determine the target parameter range allowed for the index value corresponding to the scheduling scheme according to the first optimization goal; Among the preset number of multiple optimal solutions generated latest before the iteration end condition is satisfied, use the multiple optimal solutions whose index values of the corresponding first optimization goal are within the target parameter range as the multiple candidate scheduling schemes.

[0010] In one embodiment, the step of using the simulated annealing algorithm to determine the optimal solution among the first scheduling scheme and the second scheduling scheme corresponding to the computing power resource pool information based on the first optimization goal includes: Determine whether the first scheduling scheme meets the new solution acceptance condition according to the index values of the first optimization goal corresponding to the first scheduling scheme and the third scheduling scheme respectively, where the third scheduling scheme is the previous scheduling scheme that met the new solution acceptance condition or the initial scheduling scheme; When the first scheduling scheme meets the new solution acceptance condition, determine the first scheduling scheme as the new third scheduling scheme, and determine the optimal solution according to the index values of the first optimization objective corresponding to the first scheduling scheme and the second scheduling scheme; When the first scheduling scheme does not meet the new solution acceptance condition, determine the second scheduling scheme as the optimal solution.

[0011] In one embodiment, the first optimization objective includes minimizing the scheduling cost and maximizing the scheduling speed, the index values include the scheduling cost and the scheduling speed, and the step of determining whether the first scheduling scheme meets the new solution acceptance condition according to the index values of the first optimization objective corresponding to the first scheduling scheme and the third scheduling scheme includes: When a preset condition is met, determine that the new solution acceptance condition is met, where the preset condition includes that the scheduling cost of the first scheduling scheme is less than the scheduling cost of the third scheduling scheme, or the scheduling speed of the first scheduling scheme is greater than the scheduling speed of the third scheduling scheme; When the preset condition is not met, determine the execution optimization objective between minimizing the scheduling cost and maximizing the scheduling speed according to the number of iterations in the simulated annealing algorithm, and determine whether the new solution acceptance condition is met according to the Metropolis criterion and based on the execution optimization objective.

[0012] In one embodiment, the step of determining the execution optimization objective between minimizing the scheduling cost and maximizing the scheduling speed according to the number of iterations in the simulated annealing algorithm includes: Determine the objective with the highest priority between minimizing the scheduling cost and maximizing the scheduling speed as the first objective, and determine the objective with the lowest priority between minimizing the scheduling cost and maximizing the scheduling speed as the second objective; When the number of iterations is greater than or equal to the preset number, determine the first objective as the execution optimization objective; When the number of iterations is less than the preset number, determine the second objective as the execution optimization objective.

[0013] In addition, to achieve the above object, the present application also proposes a computing power analysis device, which includes: a memory, a processor, and a computer program stored on the memory and executable on the processor, and the computer program is configured to implement the steps of the computing power analysis method as described above.

[0014] In addition, to achieve the above object, the present application also proposes a storage medium, the storage medium is a computer-readable storage medium, and a computer program is stored on the storage medium, and when the computer program is executed by a processor, it implements the steps of the computing power analysis method as described above.

[0015] One or more technical solutions proposed in this application have at least the following technical effects: First, the simulated annealing algorithm is used to analyze the computing power resource pool information corresponding to the computing power task based on the first optimization objective, and the process control parameters of the genetic algorithm are constructed from multiple candidate scheduling solutions obtained from the analysis. Then, based on the second optimization objective and the process control parameters, the genetic algorithm is used to analyze multiple candidate scheduling solutions to obtain the target scheduling solution for performing computing power scheduling. Based on this, the process of analyzing the computing power scheduling solution is no longer achieved through a single fixed algorithm, but by integrating the simulated annealing algorithm and the genetic algorithm for phased and targeted analysis, effectively improving the accuracy of computing power scheduling to precisely meet the computing power requirements of users. BRIEF DESCRIPTION OF THE DRAWINGS

[0016] The drawings here are incorporated into the specification and form a part of this specification, showing embodiments consistent with this application, and are used together with the specification to explain the principles of this application.

[0017] To more clearly illustrate the technical solutions in the embodiments of this application or the prior art, the following will briefly introduce the drawings required for use in the description of the embodiments or the prior art. Obviously, for those of ordinary skill in the art, without creative efforts, other drawings can also be obtained based on these drawings.

[0018] Figure 1 It is a schematic flowchart provided for Embodiment 1 of the computing power analysis method of this application; Figure 2 It is a schematic flowchart provided for Embodiment 2 of the computing power analysis method of this application; Figure 3 It is a schematic diagram of the device structure of the hardware operating environment involved in the computing power analysis method in the embodiments of this application.

[0019] The implementation, functional features, and advantages of the purpose of this application will be further described in combination with the embodiments and with reference to the drawings. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0020] It should be understood that the specific embodiments described here are only used to explain the technical solutions of this application and are not used to limit this application.

[0021] To better understand the technical solutions of this application, the following will be described in detail in combination with the specification drawings and specific embodiments.

[0022] The main solution of the embodiment of the present application is as follows: obtaining the computing power resource pool information corresponding to the computing power task; based on the first optimization objective, using the simulated annealing algorithm to analyze the computing power resource pool information to obtain multiple candidate scheduling schemes; constructing the process control parameters of the genetic algorithm according to the multiple candidate scheduling schemes, and based on the second optimization objective and the process control parameters, using the genetic algorithm to analyze the multiple candidate scheduling schemes to obtain the target scheduling scheme; performing computing power scheduling according to the target scheduling scheme.

[0023] In this embodiment, for the convenience of description, the computing power analysis device is used as the execution subject for elaboration below.

[0024] In the prior art, in the process of analyzing the computing power scheduling scheme, generally a single fixed algorithm is used for optimization. Such a method is limited by the limitations of the selected algorithm, resulting in the problem of poor accuracy of the analyzed computing power scheduling scheme and being unable to meet the computing power requirements of users.

[0025] The present application provides the above solution. First, the simulated annealing algorithm is used to analyze the computing power resource pool information corresponding to the computing power task based on the first optimization objective, and the process control parameters of the genetic algorithm are constructed from the multiple candidate scheduling schemes obtained by the analysis. Then, based on the second optimization objective and the process control parameters, the genetic algorithm is used to analyze the multiple candidate scheduling schemes to obtain the target scheduling scheme for performing computing power scheduling. Based on this, the process of analyzing the computing power scheduling scheme is no longer realized by a single fixed algorithm, but the simulated annealing algorithm and the genetic algorithm are integrated for phased and targeted analysis, effectively improving the accuracy of computing power scheduling to precisely meet the computing power requirements of users.

[0026] It should be noted that the execution subject of this embodiment can be a computing service device with data processing, network communication, and program running functions, such as a tablet computer, a personal computer, a mobile phone, etc., or an electronic device, a computing power analysis device, etc. that can implement the above functions. Taking the computing power analysis device as an example, this embodiment and the following embodiments will be described below.

[0027] Based on this, the embodiment of the present application provides a computing power analysis method, referring to Figure 1 , Figure 1 which is the flowchart of the first embodiment of the computing power analysis method of the present application.

[0028] In this embodiment, the computing power analysis method includes steps S10 to S40: Step S10, obtaining the computing power resource pool information corresponding to the computing power task; The computing power tasks here are the computing tasks that the computing power resources scheduled in the subsequent target scheduling scheme need to execute, such as artificial intelligence (AI) large model training tasks, image generation tasks, video generation tasks, 3D content creation tasks, and so on. The computing power resources can include cloud containers or virtual machines, etc.

[0029] Obtain the computing power resource pool information according to the task characteristic information of the computing power task (such as application scenarios, computing power demand information required for the computing power (such as computing power type, required computing power performance, performance mode, etc.), data source location distribution, etc.).

[0030] The computing power resource pool information can include multiple computing power resources available for executing the computing power task and their corresponding resource label information (for example, resource type (supercomputing power or intelligent computing power or general computing power, etc.), location information of the resources, billing information, communication speed information, computing speed information, etc.).

[0031] The computing power resources in the computing power resource pool information are all resources that meet the basic computing power requirements of the computing power task.

[0032] Step S20, based on the first optimization goal, use the simulated annealing algorithm to analyze the computing power resource pool information and obtain multiple candidate scheduling schemes; The first optimization goal can include one or more than one optimization goal. For example, the first optimization goal can include at least one of the following optimization goals: minimizing the scheduling cost, maximizing the scheduling speed, optimizing the performance, maximizing the response ability, maximizing the security, etc. The first optimization goal can be a preset fixed goal or a goal selected by the user based on their own needs.

[0033] In the process of using the simulated annealing algorithm to analyze the computing power resource pool information, based on the first optimization goal, iteratively select the best among the scheduling schemes corresponding to the computing power resource pool information. The last optimal solution obtained when meeting the iteration end condition of the simulated annealing algorithm is the scheduling scheme that best matches the first optimization goal among all the analyzed scheduling schemes.

[0034] Iteratively analyzing multiple scheduling schemes in the computing power resource pool information using the simulated annealing algorithm can be the schemes pre-screened according to a certain selection rule, or the schemes randomly combined based on multiple computing power resources in the computing power resource pool information during the iteration process.

[0035] The process of using the simulated annealing algorithm to analyze the computing power resource pool information here can be the analysis process of the traditional simulated annealing algorithm or the analysis process of the improved simulated annealing algorithm mentioned in the subsequent embodiments.

[0036] The candidate scheduling scheme can include a combination of one or more than one computing power resource. Different candidate scheduling schemes include different computing power resources or combinations of computing power resources.

[0037] The multiple candidate scheduling schemes can be the schemes generated according to the last optimal solution produced during the iterative process of the simulated annealing algorithm, or can be the schemes selected from all the optimal solutions produced during the iterative process of the simulated annealing algorithm.

[0038] Step S30: Construct the process control parameters of the genetic algorithm according to the multiple candidate scheduling schemes, and analyze the target scheduling scheme of the computing power resource pool information based on the second optimization objective and the process control parameters by using the genetic algorithm; The process control parameters can include at least one of the following: population size, crossover probability, mutation probability, selection mechanism, fitness function, termination condition, initial population, and so on.

[0039] The second optimization objective can include one or more optimization objectives. For example, the first optimization objective can include at least one of the following optimization objectives: minimizing scheduling cost, maximizing scheduling speed, optimizing performance, maximizing response ability, highest security, and so on. The second optimization objective can be a preset fixed objective or an objective selected by the user based on their own needs. In this embodiment, the first optimization objective and the second optimization objective are different objectives. In some other implementation manners, the first optimization objective and the second optimization objective can also be the same objective.

[0040] Iteratively analyze and preferentially select the multiple candidate scheduling schemes by using the genetic algorithm. The last optimal solution obtained when the termination condition of the genetic algorithm is satisfied is the scheduling scheme that best matches the second optimization objective among the multiple candidate scheduling schemes, and can be used as the target scheduling scheme.

[0041] Step S40: Execute computing power scheduling according to the target scheduling scheme.

[0042] After determining the target scheduling scheme, corresponding recommendation information can be output. The requester corresponding to the computing power task can determine whether to apply the target scheduling scheme to execute the computing power task based on the recommendation information and give feedback. When the feedback information is confirmation, the computing power resources can be scheduled according to the target scheduling scheme to execute the computing power task; when the feedback information includes non-confirmation, the negative feedback information corresponding to the scheme (such as the difference between the scheduling cost and the required cost, etc.) can be obtained, and the algorithm models corresponding to the simulated annealing algorithm and / or the genetic algorithm can be adjusted based on the negative feedback information. The simulated annealing algorithm and / or the genetic algorithm after the adjustment of the algorithm model return to execute steps S20 to S40.

[0043] This embodiment provides a computing power analysis method. First, the simulated annealing algorithm is used to analyze the computing power resource pool information corresponding to the computing power task based on the first optimization objective. The process control parameters of the genetic algorithm are constructed from the multiple candidate scheduling schemes obtained through the analysis. Then, based on the second optimization objective and the process control parameters, the genetic algorithm is used to analyze the multiple candidate scheduling schemes to obtain the target scheduling scheme for performing computing power scheduling. The process of analyzing the computing power scheduling scheme is no longer achieved through a single fixed algorithm, but rather a phased and targeted analysis is carried out by integrating the simulated annealing algorithm and the genetic algorithm, effectively improving the accuracy of computing power scheduling to precisely meet the computing power requirements of users.

[0044] In a feasible implementation manner, the step of constructing the process control parameters of the genetic algorithm according to the multiple candidate scheduling schemes includes: Determine the initial population in the genetic algorithm according to the multiple candidate scheduling schemes, and adjust the preset fitness function of the genetic algorithm corresponding to the second optimization objective according to the multiple candidate scheduling schemes to obtain the target fitness function; wherein, the process control parameters include the initial population and the target fitness function.

[0045] In this embodiment, among the multiple candidate scheduling schemes, determine the optimal solution finally generated during the iteration of the simulated annealing algorithm as the benchmark scheduling scheme, determine the index difference between the other candidate scheduling schemes and the index value of the second optimization objective corresponding to the benchmark scheduling scheme, and use the set of the benchmark scheduling scheme and a set number of candidate scheduling schemes with the smallest index difference as the initial population. For example, if the second optimization objective is to optimize the computing performance, the index difference between the performance index value of each candidate scheduling scheme other than the benchmark scheduling scheme and the performance index value of the benchmark scheduling scheme can be determined. If the number of individuals in the initial population is set to 4, then the first 3 candidate scheduling schemes with the smallest index difference and the benchmark scheduling scheme are used as the initial population.

[0046] The preset fitness function is a function preset in the genetic algorithm for evaluating the pros and cons of each individual in the population, which determines the optimization direction and effect of the genetic algorithm. In this embodiment, the preset fitness function is determined according to the second optimization objective. For example, when the second optimization objective is performance optimization, the preset fitness function can be a calculation model of the performance index value corresponding to each scheduling scheme. Based on this calculation model, the corresponding performance index value can be calculated to reflect the pros and cons of the corresponding scheduling scheme.

[0047] In this embodiment, the model parameters in the preset fitness function can be adjusted according to the index values corresponding to the multiple candidate scheduling schemes and / or the number of the multiple candidate scheduling schemes, etc., and the preset fitness function after adjusting the model parameters is used as the target fitness function.

[0048] In this embodiment, the first optimization objective includes minimizing the scheduling cost and maximizing the scheduling speed. Here, the scheduling speed can be considered as the total speed determined by comprehensively considering the communication speed and the computing speed during the use of the corresponding computing power resources. Based on this, the minimum cost among the scheduling costs corresponding to all candidate scheduling schemes is determined, and the maximum speed among the scheduling speeds corresponding to all candidate scheduling schemes is determined; The cost difference between the minimum cost and the preset minimum cost corresponding to the second optimization objective is determined, and the speed difference between the maximum speed and the preset maximum speed corresponding to the second optimization objective is determined; the preset fitness function is adjusted according to the cost difference and the speed difference to obtain the target fitness function. Among them, different second optimization objectives may correspond to different preset minimum costs and preset maximum speeds. Based on this, the first correction value of the model parameters in the preset fitness function can be determined according to the cost difference, the second correction value of the model parameters in the preset fitness function can be determined according to the speed difference, the target correction value can be determined according to the first correction value and the second correction value, and the model parameters in the preset fitness function are corrected according to the target correction value to obtain the target fitness function.

[0049] Among them, in the process of executing the genetic algorithm according to the initial population and the target fitness function here, the control parameters related to other processes involved in the genetic algorithm can be preset fixed parameters.

[0050] In this embodiment, through the above method, it can be ensured that the comprehensive capabilities achieved by the obtained target scheduling scheme in the first optimization objective and the second optimization objective can be effectively improved, further improving the accuracy of computing power scheduling, and more accurately meeting the user's computing power requirements characterized by the first optimization objective and the second optimization objective.

[0051] In other embodiments, the mutation probability in the genetic algorithm can also be determined according to the first difference between the maximum scheduling cost and the minimum scheduling cost among multiple candidate scheduling schemes and / or the second difference between the maximum scheduling speed and the minimum scheduling speed among multiple candidate scheduling schemes.

[0052] Based on any of the above embodiments, in the second embodiment of the present application, the same or similar content as the above embodiments can be referred to the above introduction and will not be repeated hereinafter. On this basis, referring to Figure 2 , step S20 includes: Step S21, based on the first optimization objective, use the simulated annealing algorithm to determine the optimal solution among the first scheduling scheme and the second scheduling scheme corresponding to the computing power resource pool information; When the iteration end condition is not satisfied, after executing step S22, return to execute step S21; when the iteration end condition is satisfied, execute step S23; In this embodiment, the first scheduling scheme may be a new solution in the simulated annealing algorithm. The second scheduling scheme may initially be the initial scheduling scheme, and the initial scheduling scheme may be the initial optimal solution in the simulated annealing algorithm. Then, it can be determined whether the new solution acceptance condition of the simulated annealing algorithm is satisfied. When the new solution acceptance condition is satisfied, it is determined whether the first scheduling scheme is the new optimal solution according to the index value of the first optimization objective corresponding to the first scheduling scheme and the index value of the second optimization objective corresponding to the second scheduling scheme. Among them, based on the second optimization objective, when the index value of the first scheduling scheme is better than that of the second scheduling scheme, it can be determined that the first scheduling scheme is the new optimal solution; when the index value of the second scheduling scheme is better than that of the first scheduling scheme, the second scheduling scheme can be maintained as the optimal solution. Among them, the new solution acceptance condition can be set according to the common new solution acceptance determination conditions in the simulated annealing algorithm, or can be comprehensively determined according to the upper limit index value or lower limit index value allowed by each optimization objective in the second optimization objective.

[0053] The iteration end condition is the condition for the simulated annealing algorithm to end the iteration during the iterative analysis of each scheduling scheme. The iteration end condition may include that the total duration of the iterative operation is greater than or equal to the preset duration, or the number of iterations is greater than or equal to the preset number of times, and so on.

[0054] Step S22: Use the current optimal solution as the new second scheduling scheme, and determine the scheduling scheme corresponding to the computing power resource pool information as the new first scheduling scheme; Here, the new scheduling scheme is a scheduling scheme different from both the first scheduling scheme and the second scheduling scheme before the current moment. The new scheduling scheme can be randomly perturbed and generated based on the simulated annealing algorithm or randomly selected from a preselected multiple scheduling schemes. Alternatively, multiple scheduling schemes corresponding to the computing power resource pool information can be pre-generated, and the generated multiple scheduling schemes are sorted according to the first optimization objective to obtain a preset order. Then, the scheduling scheme ranked first in the preset order is the initial first scheduling scheme, the scheduling scheme ranked second in the preset order is the initial second scheduling scheme, and the scheduling scheme ranked first among all the remaining scheduling schemes after excluding the previous first scheduling scheme and the second scheduling scheme in the preset order is used as the new first scheduling scheme. For example, when the priority of the scheduling cost in the first optimization objective is higher than the priority of the scheduling speed, the preset order is obtained by sorting the multiple scheduling schemes corresponding to the scheduling cost from high to low; when the priority of the scheduling speed in the first optimization objective is higher than the priority of the scheduling cost, the preset order is obtained by sorting the multiple scheduling schemes corresponding to the scheduling speed from low to high.

[0055] Step S23: Determine the multiple candidate scheduling schemes from the latest generated preset number of optimal solutions before the iteration end condition is satisfied.

[0056] In this embodiment, among a preset number of the most recent optimal solutions generated before the iteration end condition is satisfied, multiple optimal solutions that meet the preset conditions can be selected as multiple candidate scheduling schemes, or all the optimal solutions can be used as multiple candidate scheduling schemes.

[0057] In this embodiment, during the iterative analysis process of the simulated annealing algorithm, multiple optimal solutions newly generated can be considered as multiple scheduling schemes closest to the first optimization goal. These are used as alternative scheduling schemes and further analyzed through the subsequent genetic algorithm to obtain the target scheduling scheme, which is beneficial to further improving the accuracy of computing power scheduling.

[0058] In a feasible implementation manner, the step of determining the multiple candidate scheduling schemes among a preset number of the most recent optimal solutions generated before the iteration end condition is satisfied includes: determining the target parameter range allowed for the index value corresponding to the scheduling scheme according to the first optimization goal; among a preset number of the most recent optimal solutions generated before the iteration end condition is satisfied, taking the multiple optimal solutions whose index values of the corresponding first optimization goal are within the target parameter range as the multiple candidate scheduling schemes.

[0059] When the first optimization goal includes minimizing the scheduling cost, the maximum scheduling cost set by the user can be obtained, and the cost interval less than or equal to the maximum scheduling cost is used as the corresponding target parameter range; and / or, when the first optimization goal includes maximizing the scheduling speed, the minimum scheduling speed set by the user can be obtained, and the speed interval greater than or equal to the minimum scheduling speed is used as the corresponding target parameter range.

[0060] When the first optimization goal includes minimizing the scheduling cost and maximizing the scheduling speed, the optimal solution is used as a candidate scheduling scheme when both the scheduling cost and the scheduling speed corresponding to the optimal solution are within the corresponding target parameter ranges, and the optimal solution is not used as a candidate scheduling scheme when either the scheduling cost or the scheduling speed corresponding to the optimal solution is not within the corresponding target parameter ranges.

[0061] In this embodiment, through the above method, it is beneficial to further improve the accuracy of the subsequent obtained target scheduling scheme.

[0062] In a feasible implementation manner, the step of determining the optimal solution among the first scheduling scheme and the second scheduling scheme corresponding to the computing power resource pool information based on the first optimization objective includes: determining whether the first scheduling scheme meets the new solution acceptance condition according to the index values of the first optimization objective corresponding to the first scheduling scheme and the third scheduling scheme, where the third scheduling scheme is the previous scheduling scheme that meets the new solution acceptance condition or the initial scheduling scheme; when the first scheduling scheme meets the new solution acceptance condition, determining the first scheduling scheme as the new third scheduling scheme, and determining the optimal solution according to the index values of the first optimization objective corresponding to the first scheduling scheme and the second scheduling scheme; when the first scheduling scheme does not meet the new solution acceptance condition, determining the second scheduling scheme as the optimal solution.

[0063] Among them, when there is no previous scheduling scheme that meets the new solution acceptance condition, the third scheduling scheme is the initial scheduling scheme; when there is a previous scheduling scheme that meets the new solution acceptance condition, the third scheduling scheme is the previous scheduling scheme that meets the new solution acceptance condition.

[0064] When the first optimization objective includes minimizing the scheduling cost, the index value may include the scheduling cost; when the first optimization objective includes maximizing the scheduling speed, the index value includes the scheduling speed.

[0065] In this embodiment, the first optimization objective includes minimizing the scheduling cost and maximizing the scheduling speed, and the index values include the scheduling cost and the scheduling speed. Then the process of determining whether the first scheduling scheme meets the new solution acceptance condition is as follows: when the preset condition is met, it is determined that the new solution acceptance condition is met, where the preset condition includes that the scheduling cost of the first scheduling scheme is less than the scheduling cost of the third scheduling scheme, or the scheduling speed of the first scheduling scheme is greater than the scheduling speed of the third scheduling scheme; when the preset condition is not met, the execution optimization objective between minimizing the scheduling cost and maximizing the scheduling speed is determined according to the iteration number in the simulated annealing algorithm, and whether the new solution acceptance condition is met is determined according to the Metropolis criterion (the Chinese name is "Metropolis criterion", which is a kind of acceptance-rejection criterion, and its role is to find the lowest energy state on the energy surface). Among them, each process of determining the optimal solution can be regarded as an iteration, and the iteration number is incremented by one. Here, the corresponding execution optimization objectives are different in different intervals where the execution number is located. The energy difference between the two schemes can be determined according to the index values of the execution optimization objectives corresponding to the first scheduling scheme and the third scheduling scheme, and whether the new solution acceptance condition is met is determined based on the energy difference and the Metropolis criterion.

[0066] In this embodiment, the first optimization objective includes minimizing the scheduling cost and maximizing the scheduling speed. The metric values include the scheduling cost and the scheduling speed. In the process of determining the optimal solution, a first probability can be randomly generated, and a first probability threshold can be generated according to the number of iterations. When the first probability is less than or equal to the first probability threshold and the scheduling cost corresponding to the first scheduling scheme is less than the scheduling cost corresponding to the second scheduling scheme, the first scheduling scheme can be determined as the optimal solution; when the first probability is greater than the first probability threshold and the scheduling speed corresponding to the first scheduling scheme is greater than the scheduling speed corresponding to the second scheduling scheme, the first scheduling scheme is determined as the optimal solution; when the first probability is less than or equal to the first probability threshold and the scheduling cost corresponding to the first scheduling scheme is greater than or equal to the scheduling cost corresponding to the second scheduling scheme, the second scheduling scheme is determined as the optimal solution; when the first probability is greater than the first probability threshold and the scheduling speed corresponding to the first scheduling scheme is less than or equal to the scheduling speed corresponding to the second scheduling scheme, the second scheduling scheme is determined as the optimal solution.

[0067] In this embodiment, through the above method, it is beneficial to ensure that the target scheduling scheme can take into account the first optimization objective and the second optimization objective at the same time, and further improve the accuracy of computing power scheduling. Among them, when the first optimization objective includes the scheduling cost and the scheduling speed, the corresponding execution optimization objective is determined based on the number of executions to judge whether the new solution acceptance condition is met, so as to realize the optimization of different objectives in stages, ensure that the obtained target scheduling scheme can take into account different optimization objectives, and further improve the accuracy of computing power scheduling.

[0068] In other embodiments, when the first optimization objective includes minimizing the scheduling cost, the first scheduling scheme can be determined as the optimal solution when the scheduling cost of the first scheduling scheme is less than the scheduling cost of the second scheduling scheme, otherwise the second scheduling scheme is determined as the optimal solution; when the first optimization objective includes maximizing the scheduling speed, the first scheduling scheme can be determined as the optimal solution when the scheduling cost of the first scheduling scheme is greater than the scheduling cost of the second scheduling scheme, otherwise the second scheduling scheme is determined as the optimal solution.

[0069] In a feasible implementation manner, the step of determining the execution optimization objective between minimizing the scheduling cost and maximizing the scheduling speed according to the number of iterations in the simulated annealing algorithm includes: determining the objective with the highest priority between minimizing the scheduling cost and maximizing the scheduling speed as the first objective, and determining the objective with the lowest priority between minimizing the scheduling cost and maximizing the scheduling speed as the second objective; when the number of iterations is greater than or equal to the preset number of times, determining the first objective as the execution optimization objective; when the number of iterations is less than the preset number of times, determining the second objective as the execution optimization objective.

[0070] The priority between the two objectives in the first optimization objective here can be selected by the user.

[0071] Based on this, in the initial stage of iteration, the solution can be optimized with the lowest-priority objective first, and then gradually converge to the solution corresponding to the highest-priority objective in the later stage of iteration, so as to ensure that the trade-off effect between the scheduling cost and the scheduling speed of the target scheduling solution can accurately meet the user requirements.

[0072] It should be noted that the above examples are only for understanding this application and do not constitute a limitation on the computing power analysis method of this application. Based on this technical concept, more forms of simple transformations are within the protection scope of this application.

[0073] Referring to Figure 3 , this application provides a computing power analysis device 1, which may include: at least one processor 1001; and a memory 1002, an input / output (I / O) interface 1003, etc. that are communicatively connected to the at least one processor 1001; wherein, the memory 1002 stores instructions executable by the at least one processor 1001, and the instructions are executed by the at least one processor 1001 to enable the at least one processor 1001 to execute the computing power analysis method in the first embodiment above.

[0074] Next, referring to Figure 3 , which shows a schematic structural diagram of the computing power analysis device 1 suitable for implementing the embodiments of this application. The computing power analysis device 1 in the embodiments of this application may include, but is not limited to, mobile terminals such as mobile phones, laptop computers, etc. and fixed terminals such as digital TVs, desktop computers, etc. Figure 3 The shown computing power analysis device 1 is only an example and should not bring any limitation to the functions and usage scope of the embodiments of this application.

[0075] As Figure 3As shown, the computing power analysis device 1 may include a processor 1001 (such as a central processing unit, a graphics processing unit, etc.), which may perform various appropriate actions and processes according to the programs stored in the memory 1002. Here, the programs in the memory 1002 may be programs in a read-only memory (ROM: Read Only Memory) or programs loaded from a storage device into a random access memory (RAM: Random Access Memory). In the RAM, various programs and data required for the operation of the computing power analysis device 1 are also stored. The processor 1001, the memory 1002 (ROM and RAM), the input / output (I / O) interface 1003, etc. are connected to each other through a communication bus 1007. Generally, the following systems may be connected to the I / O interface 1003: an input device 1004 including, for example, a touch screen, a touchpad, a keyboard, a mouse, an image sensor, a microphone, an accelerometer, a gyroscope, etc.; an output device 1005 including, for example, a liquid crystal display (LCD: Liquid Crystal Display), a speaker, a vibrator, etc.; and a communication device 1006, etc. The communication device 1006 may allow the computing power analysis device 1 to communicate with other devices wirelessly or wiredly to exchange data. Although the computing power analysis device 1 with various devices is shown in the figure, it should be understood that it is not required to implement or have all the shown devices. More or fewer devices may be implemented or had alternatively.

[0076] In particular, according to the embodiments disclosed in the present application, the method flow described in the above embodiments may be implemented as a computer software program. For example, the embodiments disclosed in the present application include a computer program product, which includes a computer program carried on a computer-readable medium, and the computer program contains program codes for executing the method shown in the flowchart. In such an embodiment, the computer program may be downloaded and installed from the network through the communication device, or installed from the memory 1002. When the computer program is executed by the processor 1001, the above functions defined in the computing power analysis method of the embodiments disclosed in the present application are executed.

[0077] The computing power analysis device provided by the present application adopts the computing power analysis method in the above embodiments, and can solve the technical problem of how to improve the accuracy of computing power scheduling to accurately meet the computing power requirements of users. Compared with the prior art, the beneficial effects of the computing power analysis device provided by the present application are the same as those of the computing power analysis method provided by the above embodiments, and the other technical features in this computing power analysis device are the same as the features disclosed in the method of the previous embodiment, and will not be elaborated here.

[0078] The present application provides a computer-readable storage medium, having computer-readable program instructions (i.e., a computer program) stored thereon, and the computer-readable program instructions are used to execute the computing power analysis method in the above embodiments.

[0079] The computer-readable storage medium provided by the present application may be, for example, a USB flash drive, but is not limited to electrical, magnetic, optical, electromagnetic, infrared, or semiconductor systems, devices, or components, or any combination of the above. More specific examples of computer-readable storage media may include, but are not limited to: electrical connections with one or more wires, portable computer disks, hard disks, random access memory (RAM), read-only memory (ROM), erasable programmable read-only memory (EPROM) or flash memory, optical fibers, portable compact disk read-only memory (CD-ROM), optical storage devices, magnetic storage devices, or any suitable combination of the above. In this embodiment, the computer-readable storage medium may be any tangible medium that contains or stores a program that can be used by or in conjunction with an instruction execution system, device, or component. The program code contained on the computer-readable storage medium may be transmitted using any appropriate medium, including but not limited to: wires, optical cables, RF (radio frequency), etc., or any suitable combination of the above.

[0080] The above computer-readable storage medium may be included in the computing power analysis device; or it may exist independently and not be assembled into the computing power analysis device.

[0081] The above computer-readable storage medium carries one or more programs. When the one or more programs are executed by the computing power analysis device, the computing power analysis device is caused to execute the following processes: obtaining computing power resource pool information corresponding to a computing power task; based on a first optimization objective, using the simulated annealing algorithm to analyze the computing power resource pool information to obtain multiple candidate scheduling schemes; constructing process control parameters of a genetic algorithm according to the multiple candidate scheduling schemes, and based on a second optimization objective and the process control parameters, using the genetic algorithm to analyze the multiple candidate scheduling schemes to obtain a target scheduling scheme; and performing computing power scheduling according to the target scheduling scheme.

[0082] Computer program code for performing the operations of this application can be written in one or more programming languages or combinations thereof. The above-mentioned programming languages include object-oriented programming languages such as Java, Smalltalk, C++, and also include conventional procedural programming languages such as the "C" language or similar programming languages. The program code can be executed entirely on the user's computer, partially on the user's computer, executed as an independent software package, partially on the user's computer and partially on a remote computer, or entirely on a remote computer or server. In the case of a remote computer, the remote computer can be connected to the user's computer through any type of network, including a local area network (LAN) or a wide area network (WAN), or it can be connected to an external computer (for example, by connecting through the Internet using an Internet service provider).

[0083] The readable storage medium provided by this application is a computer-readable storage medium. The computer-readable storage medium stores computer-readable program instructions (i.e., computer programs) for performing the above-mentioned computing power analysis method, and can solve the technical problem of how to improve the accuracy of computing power scheduling to precisely meet the computing power requirements of users. Compared with the prior art, the beneficial effects of the computer-readable storage medium provided by this application are the same as those of the computing power analysis method provided in the above embodiments, and will not be elaborated here.

[0084] The flowcharts and block diagrams in the accompanying drawings illustrate the possible architectures, functions, and operations of systems, methods, and computer program products according to various embodiments of this application. In this regard, each block in the flowchart or block diagram may represent a module, a program segment, or a part of the code, and this module, program segment, or part of the code contains one or more executable instructions for implementing the specified logical function. It should also be noted that in some alternative implementations, the functions marked in the blocks may occur in a different order than that marked in the accompanying drawings. For example, two consecutive blocks shown may actually be executed substantially in parallel, and they may sometimes be executed in the reverse order, depending on the functions involved. It should also be noted that each block in the block diagram and / or flowchart, as well as the combination of blocks in the block diagram and / or flowchart, can be implemented by a dedicated hardware-based system for performing the specified functions or operations, or can be implemented by a combination of dedicated hardware and computer instructions.

[0085] It should be understood that each part disclosed in this application can be implemented by hardware, software, firmware, or a combination thereof. The modules described in the embodiments of this application can be implemented in software or in hardware. Among them, the name of the module does not constitute a limitation to the unit itself in some cases. In the description of the above embodiments, specific features, structures, materials, or characteristics can be combined in a suitable manner in any one or more embodiments or examples.

[0086] The above are only some embodiments of this application, and thus do not limit the patent scope of this application. Any equivalent structural transformation made under the technical concept of this application by using the content of the specification and drawings of this application, or any direct / indirect application in other related technical fields, is included in the patent protection scope of this application. Therefore, the protection scope of this application shall be subject to the protection scope of the claims.

Claims

1. A computing power analysis method, characterized in that, The method described above includes: Obtaining the computing power resource pool information corresponding to the computing power task; Based on the first optimization objective, using the simulated annealing algorithm to analyze the computing power resource pool information to obtain multiple candidate scheduling schemes; Constructing the process control parameters of the genetic algorithm according to the multiple candidate scheduling schemes, and based on the second optimization objective and the process control parameters, using the genetic algorithm to analyze the multiple candidate scheduling schemes to obtain the target scheduling scheme; Performing computing power scheduling according to the target scheduling scheme.

2. The computing power analysis method according to claim 1, wherein The step of constructing the process control parameters of the genetic algorithm according to the multiple candidate scheduling schemes includes: Determining the initial population in the genetic algorithm according to the multiple candidate scheduling schemes, and adjusting the preset fitness function of the genetic algorithm corresponding to the second optimization objective according to the multiple candidate scheduling schemes to obtain the target fitness function; Wherein, the process control parameters include the initial population and the target fitness function.

3. The computing power analysis method according to claim 2, wherein The first optimization objective includes minimizing the scheduling cost and maximizing the scheduling speed. The step of adjusting the preset fitness function of the genetic algorithm corresponding to the second optimization objective according to the multiple candidate scheduling schemes to obtain the target fitness function includes: Determining the minimum cost among the scheduling costs corresponding to all candidate scheduling schemes, and determining the maximum speed among the scheduling speeds corresponding to all candidate scheduling schemes; Determining the cost difference between the minimum cost and the preset minimum cost corresponding to the second optimization objective, and determining the speed difference between the maximum speed and the preset maximum speed corresponding to the second optimization objective; Adjusting the preset fitness function according to the cost difference and the speed difference to obtain the target fitness function.

4. The computing power analysis method according to any one of claims 1 to 3, characterized in that The step of using the simulated annealing algorithm to analyze the computing power resource pool information based on the first optimization objective to obtain multiple candidate scheduling schemes includes: Based on the first optimization objective, using the simulated annealing algorithm to determine the optimal solution among the first scheduling scheme and the second scheduling scheme corresponding to the computing power resource pool information; When the iteration end condition is not satisfied, taking the current optimal solution as the new second scheduling scheme, determining the new scheduling scheme in the computing power resource pool information as the new first scheduling scheme, and returning to execute the step of using the simulated annealing algorithm to determine the optimal solution among the first scheduling scheme and the second scheduling scheme corresponding to the computing power resource pool information based on the first optimization objective; When the iteration end condition is satisfied, determining the multiple candidate scheduling schemes from the preset number of multiple optimal solutions generated latest before the iteration end condition is satisfied.

5. The computing power analysis method according to claim 4, wherein The step of determining the multiple candidate scheduling schemes from the preset number of multiple optimal solutions generated latest before the iteration end condition is satisfied includes: Determining the target parameter range allowed for the index value corresponding to the scheduling scheme according to the first optimization objective; Among the preset number of multiple optimal solutions generated latest before the iteration end condition is satisfied, taking the multiple optimal solutions whose index values of the corresponding first optimization objective are within the target parameter range as the multiple candidate scheduling schemes.

6. The computing power analysis method according to claim 4, wherein The step of determining the optimal solution in the first scheduling plan and the second scheduling plan corresponding to the computing power resource pool information based on the first optimization goal includes: Determine whether the first scheduling plan meets the new solution acceptance condition according to the index values of the first optimization goal corresponding to the first scheduling plan and the third scheduling plan, where the third scheduling plan is the previous scheduling plan that meets the new solution acceptance condition or the initial scheduling plan; When the first scheduling plan meets the new solution acceptance condition, determine the first scheduling plan as the new third scheduling plan, and determine the optimal solution according to the index values of the first optimization goal corresponding to the first scheduling plan and the second scheduling plan; When the first scheduling plan does not meet the new solution acceptance condition, determine the second scheduling plan as the optimal solution.

7. The computing power analysis method according to claim 6, wherein The first optimization goal includes minimizing the scheduling cost and maximizing the scheduling speed. The index values include the scheduling cost and the scheduling speed. The step of determining whether the first scheduling plan meets the new solution acceptance condition according to the index values of the first optimization goal corresponding to the first scheduling plan and the third scheduling plan includes: When a preset condition is met, determine that the new solution acceptance condition is met, where the preset condition includes that the scheduling cost of the first scheduling plan is less than the scheduling cost of the third scheduling plan, or the scheduling speed of the first scheduling plan is greater than the scheduling speed of the third scheduling plan; When the preset condition is not met, determine the execution optimization goal between minimizing the scheduling cost and maximizing the scheduling speed according to the number of iterations in the simulated annealing algorithm, and determine whether the new solution acceptance condition is met according to the Metropolis criterion and based on the execution optimization goal.

8. The computing power analysis method according to claim 7, wherein The step of determining the execution optimization goal between minimizing the scheduling cost and maximizing the scheduling speed according to the number of iterations in the simulated annealing algorithm includes: Determine the goal with the highest priority between minimizing the scheduling cost and maximizing the scheduling speed as the first goal, and determine the goal with the lowest priority between minimizing the scheduling cost and maximizing the scheduling speed as the second goal; When the number of iterations is greater than or equal to the preset number of times, determine the first goal as the execution optimization goal; When the number of iterations is less than the preset number of times, determine the second goal as the execution optimization goal.

9. A computing power analysis device, characterized in that, The computing power analysis device includes: a memory, a processor, and a computer program stored on the memory and executable on the processor, and the computer program is configured to implement the steps of the computing power analysis method according to any one of claims 1 to 8.

10. A storage medium, characterized in that, The storage medium is a computer-readable storage medium, and a computer program is stored on the storage medium. When the computer program is executed by a processor, it implements the steps of the computing power analysis method according to any one of claims 1 to 8.

Images