Computing power scheduling method and system based on multi-model collaboration

Abstract

The invention discloses a computing power scheduling method and system based on multi-model cooperation. The method comprises the following steps: acquiring a plurality of computing power requests of a plurality of computing power users; determining a request parameter corresponding to each computing power request according to the request data of each computing power request; respectively inputting request parameters corresponding to all the computing power requests into prediction models corresponding to a plurality of computing power index types to obtain a plurality of corresponding predicted computing power index parameters; and calculating a computing power resource scheduling strategy corresponding to the plurality of computing power requests according to a dynamic programming algorithm based on a target function and a limiting condition associated with the plurality of predicted computing power index parameters. Visibly, the computing power scheduling method and system can realize computing power scheduling for computing power requests of multiple users more efficiently and reasonably, improve the efficiency and effect of computing power resource scheduling, and provide better user experience for computing power leasing users.

Description

Technical Field

[0001] The present invention relates to the field of data processing technology, and in particular to a computing power scheduling method and system based on multi-model collaboration. Background Art

[0002] With the increasing demand for big data computing and the widespread improvement of data processing capabilities, users' demand for leasing computing power is also increasing. The computing power leasing platform begins to need to deal with the problem of computing power resource scheduling between a large number of users' computing power requests. Among them, how to improve the accuracy and efficiency of scheduling has become an important technical issue. When implementing the scheduling of computing power resources, the existing technology is still mostly based on the parameters of computing power equipment and simple data judgment rules. It does not consider the combination of prediction algorithms and dynamic programming algorithms corresponding to indicators. The efficiency and rationality of its scheduling cannot meet the requirements. It can be seen that the existing technology has defects and needs to be solved urgently. Summary of the invention

[0003] The technical problem to be solved by the present invention is to provide a computing power scheduling method and system based on multi-model collaboration, which can more efficiently and reasonably realize computing power scheduling for computing power requests of multiple users, improve the efficiency and effect of computing power resource scheduling, and provide a better user experience for computing power leasing users.

[0004] In order to solve the above technical problems, the first aspect of the present invention discloses a computing power scheduling method based on multi-model collaboration, the method comprising: Get multiple computing power requests from multiple computing power users; Determine, according to the request data of each computing power request, a request parameter corresponding to each computing power request; Inputting the request parameters corresponding to all the computing power requests into the prediction models corresponding to the multiple computing power indicator types respectively to obtain the corresponding multiple predicted computing power indicator parameters; Based on the objective functions and constraints associated with the multiple predicted computing power indicator parameters, a computing power resource scheduling strategy corresponding to the multiple computing power requests is calculated according to a dynamic programming algorithm.

[0005] As an optional implementation, in the first aspect of the present invention, the request data includes computing power user information, request data, request device information, request network environment information and request location information.

[0006] As an optional implementation, in the first aspect of the present invention, the request parameters include the requested task type, the requested task urgency, the requested personnel level, the requested task industry category and the requested malicious possibility parameter.

[0007] As an optional implementation, in the first aspect of the present invention, determining the request parameter corresponding to each computing power request according to the request data of each computing power request includes: For each computing power request, obtaining request data corresponding to the computing power request; The request data is input into a neural network model corresponding to a plurality of request parameter types to obtain request parameters corresponding to the computing power request; the request parameter types are request task type, request task urgency, request personnel level, request task industry category or request malicious possibility parameter.

[0008] As an optional embodiment, in the first aspect of the present invention, the neural network model is an RNN neural network model, which is trained by a training data set including request parameter annotations of multiple training request data and corresponding request parameter types, and optimizes model parameters based on a gradient descent algorithm and a cross entropy loss function until convergence.

[0009] As an optional implementation, in the first aspect of the present invention, the predicted computing power indicator parameter is the computing power task energy consumption requirement, the computing power task time limit, the computing power task performance requirement or the computing power device type requirement.

[0010] As an optional implementation, in the first aspect of the present invention, the computing resource scheduling strategy corresponding to the multiple computing power requests is calculated based on the objective function and constraint conditions associated with the multiple predicted computing power indicator parameters according to the dynamic programming algorithm, including: Screening out multiple target computing devices that meet the computing device type requirements from multiple candidate computing devices; Obtaining device parameters corresponding to each of the target computing power devices; the device parameters include device computing performance, device energy consumption, and device working stability; Determine the objective function and constraint conditions according to the energy consumption requirement of the computing task, the time limit of the computing task, the performance requirement of the computing task, and the device parameters corresponding to each of the target computing devices; Based on the objective function and the constraints, the computing power scheduling schemes corresponding to all the target computing power devices and all the computing power requests are iteratively calculated according to the dynamic programming algorithm until the optimal computing power scheduling scheme is obtained as the computing power resource scheduling strategy corresponding to the multiple computing power requests.

[0011] As an optional implementation, in the first aspect of the present invention, determining the objective function and the constraint conditions according to the energy consumption requirement of the computing task, the time limit of the computing task and the performance requirement of the computing task, and the device parameters corresponding to each of the target computing devices includes: Determine the objective function so that the number of computing power requests allocated to each of the target computing power devices is minimized; Identify constraints including: The ratio between the first product value of the number of all computing power requests allocated to each of the target computing power devices and the computing power task energy consumption requirement and the device energy consumption corresponding to the target computing power device is less than a first ratio threshold; The ratio between the number of all computing power requests allocated to each of the target computing power devices and the second product value of the computing power task time limit and the working stability corresponding to the target computing power device is less than a second ratio threshold; The ratio between the third product value of the number of all computing power requests allocated to each of the target computing power devices and the computing power task performance requirement and the device computing performance corresponding to the target computing power device is less than a third ratio threshold.

[0012] A second aspect of an embodiment of the present invention discloses a computing power scheduling system based on multi-model collaboration, the system comprising: An acquisition module, used to acquire multiple computing power requests from multiple computing power users; A determination module, used to determine the request parameters corresponding to each computing power request according to the request data of each computing power request; A prediction module, used to input the request parameters corresponding to all the computing power requests into the prediction models corresponding to the multiple computing power indicator types, so as to obtain the corresponding multiple predicted computing power indicator parameters; The scheduling module is used to calculate the computing power resource scheduling strategy corresponding to the multiple computing power requests based on the objective function and constraints associated with the multiple predicted computing power indicator parameters and according to the dynamic programming algorithm.

[0013] As an optional implementation, in the second aspect of the present invention, the request data includes computing power user information, request data, request device information, request network environment information and request location information.

[0014] As an optional implementation, in the second aspect of the present invention, the request parameters include the requested task type, the requested task urgency, the requested personnel level, the requested task industry category and the requested malicious possibility parameter.

[0015] As an optional implementation, in the second aspect of the present invention, the specific manner in which the determination module determines the request parameters corresponding to each computing power request according to the request data of each computing power request includes: For each computing power request, obtaining request data corresponding to the computing power request; The request data is input into a neural network model corresponding to a plurality of request parameter types to obtain request parameters corresponding to the computing power request; the request parameter types are request task type, request task urgency, request personnel level, request task industry category or request malicious possibility parameter.

[0016] As an optional embodiment, in the second aspect of the present invention, the neural network model is an RNN neural network model, which is trained by a training data set including request parameter annotations of multiple training request data and corresponding request parameter types, and optimizes model parameters based on a gradient descent algorithm and a cross entropy loss function until convergence.

[0017] As an optional implementation, in the second aspect of the present invention, the predicted computing power indicator parameter is the computing power task energy consumption requirement, the computing power task time limit, the computing power task performance requirement or the computing power device type requirement.

[0018] As an optional implementation, in the second aspect of the present invention, the scheduling module calculates the specific manner of the computing power resource scheduling strategy corresponding to the multiple computing power requests based on the objective function and constraint conditions associated with the multiple predicted computing power indicator parameters according to the dynamic programming algorithm, including: Screening out multiple target computing devices that meet the computing device type requirements from multiple candidate computing devices; Obtaining device parameters corresponding to each of the target computing power devices; the device parameters include device computing performance, device energy consumption, and device working stability; Determine the objective function and constraint conditions according to the energy consumption requirement of the computing task, the time limit of the computing task, the performance requirement of the computing task, and the device parameters corresponding to each of the target computing devices; Based on the objective function and the constraints, the computing power scheduling schemes corresponding to all the target computing power devices and all the computing power requests are iteratively calculated according to the dynamic programming algorithm until the optimal computing power scheduling scheme is obtained as the computing power resource scheduling strategy corresponding to the multiple computing power requests.

[0019] As an optional implementation, in the second aspect of the present invention, the scheduling module determines the specific manner of the objective function and the constraint conditions according to the energy consumption requirement of the computing task, the time limit of the computing task and the performance requirement of the computing task, and the device parameters corresponding to each of the target computing devices, including: Determine the objective function so that the number of computing power requests allocated to each of the target computing power devices is minimized; Identify constraints including: The ratio between the first product value of the number of all computing power requests allocated to each of the target computing power devices and the computing power task energy consumption requirement and the device energy consumption corresponding to the target computing power device is less than a first ratio threshold; The ratio between the number of all computing power requests allocated to each of the target computing power devices and the second product value of the computing power task time limit and the working stability corresponding to the target computing power device is less than a second ratio threshold; The ratio between the third product value of the number of all computing power requests allocated to each of the target computing power devices and the computing power task performance requirement and the device computing performance corresponding to the target computing power device is less than a third ratio threshold.

[0020] The third aspect of the present invention discloses another computing power scheduling system based on multi-model collaboration, the system comprising: A memory storing executable program code; a processor coupled to the memory; The processor calls the executable program code stored in the memory to execute part or all of the steps in the computing power scheduling method based on multi-model collaboration disclosed in the first aspect of the present invention.

[0021] The fourth aspect of the present invention discloses a computer storage medium, which stores computer instructions. When the computer instructions are called, they are used to execute part or all of the steps in the computing power scheduling method based on multi-model collaboration disclosed in the first aspect of the present invention.

[0022] Compared with the prior art, the embodiments of the present invention have the following beneficial effects: The present invention can determine the corresponding request parameters according to the request data of each computing power request, and then predict the corresponding multiple predicted computing power indicator parameters based on the prediction models corresponding to the multiple computing power indicator types, so as to calculate a reasonable and efficient computing power resource scheduling strategy based on the objective function and constraint conditions associated with the predicted computing power indicator parameters, thereby being able to more efficiently and reasonably implement computing power scheduling for the computing power requests of multiple users, improve the efficiency and effect of computing power resource scheduling, and provide a better user experience for computing power leasing users. BRIEF DESCRIPTION OF THE DRAWINGS

[0023] In order to more clearly illustrate the technical solutions in the embodiments of the present invention, the following briefly introduces the drawings required for use in the description of the embodiments. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without creative work.

[0024] Figure 1It is a flow chart of a computing power scheduling method based on multi-model collaboration disclosed in an embodiment of the present invention.

[0025] Figure 2 It is a structural diagram of a computing power scheduling system based on multi-model collaboration disclosed in an embodiment of the present invention.

[0026] Figure 3 It is a structural diagram of another computing power scheduling system based on multi-model collaboration disclosed in an embodiment of the present invention. DETAILED DESCRIPTION

[0027] In order to enable those skilled in the art to better understand the scheme of the present invention, the technical scheme in the embodiments of the present invention will be clearly and completely described below in conjunction with the drawings in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0028] The terms "first", "second", etc. in the specification and claims of the present invention and the above-mentioned drawings are used to distinguish different objects, rather than to describe a specific order. In addition, the terms "including" and "having" and any variations thereof are intended to cover non-exclusive inclusions. For example, a process, method, device, product or equipment that includes a series of steps or units is not limited to the listed steps or units, but may optionally include steps or units that are not listed, or may optionally include other steps or units that are inherent to these processes, methods, products or equipment.

[0029] Reference to "embodiments" herein means that a particular feature, structure, or characteristic described in conjunction with the embodiments may be included in at least one embodiment of the present invention. The appearance of the phrase in various places in the specification does not necessarily refer to the same embodiment, nor is it an independent or alternative embodiment that is mutually exclusive with other embodiments. It is explicitly and implicitly understood by those skilled in the art that the embodiments described herein may be combined with other embodiments.

[0030] The present invention discloses a computing power scheduling method and system based on multi-model collaboration, which can determine the corresponding request parameters according to the request data of each computing power request, and then predict the corresponding multiple predicted computing power index parameters based on the prediction models corresponding to the multiple computing power index types, so as to calculate a reasonable and efficient computing power resource scheduling strategy based on the objective function and constraint conditions associated with the predicted computing power index parameters, so as to realize computing power scheduling for the computing power requests of multiple users more efficiently and reasonably, improve the efficiency and effect of computing power resource scheduling, and provide a better user experience for computing power leasing users. The following are detailed descriptions.

[0031] Embodiment 1 See also Figure 1 , Figure 1 It is a flowchart of a computing power scheduling method based on multi-model collaboration disclosed in an embodiment of the present invention. Figure 1 The described computing power scheduling method based on multi-model collaboration can be applied to a data processing system / data processing device / data processing server (wherein the server includes a local processing server or a cloud processing server). Figure 1 As shown, the computing power scheduling method based on multi-model collaboration may include the following operations: 101. Obtain multiple computing power requests from multiple computing power users.

[0032] 102. Determine request parameters corresponding to each computing power request based on request data of each computing power request. 103. Input the request parameters corresponding to all computing power requests into the prediction models corresponding to the multiple computing power indicator types respectively to obtain the corresponding multiple predicted computing power indicator parameters. 104. Based on the objective function and constraints associated with multiple predicted computing power indicator parameters, a computing power resource scheduling strategy corresponding to multiple computing power requests is calculated according to a dynamic programming algorithm.

[0033] It can be seen that the above-mentioned embodiments of the invention can determine the corresponding request parameters according to the request data of each computing power request, and then predict the corresponding multiple predicted computing power indicator parameters based on the prediction models corresponding to the multiple computing power indicator types, so as to calculate a reasonable and efficient computing power resource scheduling strategy based on the objective function and constraint conditions associated with the predicted computing power indicator parameters, thereby being able to more efficiently and reasonably implement computing power scheduling for the computing power requests of multiple users, improve the efficiency and effect of computing power resource scheduling, and provide a better user experience for computing power leasing users.

[0034] As an optional embodiment, in the above steps, the request data includes computing power user information, request data, request device information, request network environment information and request location information.

[0035] It can be seen that through the above optional embodiments, the content of the request data is clarified, and the data content of the user's computing power request can be comprehensively characterized to facilitate subsequent computing power resource scheduling, assist in achieving more efficient and reasonable computing power scheduling for multiple users' computing power requests, improve the efficiency and effectiveness of computing power resource scheduling, and provide a better user experience for computing power leasing users.

[0036] As an optional embodiment, in the above steps, the request parameters include the requested task type, the requested task urgency, the requested personnel level, the requested task industry category and the requested malicious possibility parameter.

[0037] It can be seen that through the above optional embodiments, the content of the request parameters is clarified, which can comprehensively characterize the characteristics of the user's computing power request, so as to facilitate the subsequent computing power resource scheduling, and assist in realizing more efficient and reasonable computing power scheduling for multiple users' computing power requests, thereby improving the efficiency and effectiveness of computing power resource scheduling, and providing a better user experience for computing power leasing users.

[0038] As an optional embodiment, in the above steps, determining the request parameters corresponding to each computing power request according to the request data of each computing power request includes: For each computing power request, obtain the request data corresponding to the computing power request; The request data is input into a neural network model corresponding to multiple request parameter types to obtain the request parameters corresponding to the computing power request; the request parameter types are the request task type, the request task urgency, the request personnel level, the request task industry category or the request malicious possibility parameter.

[0039] It can be seen that through the above-mentioned optional embodiments, it is possible to predict the request parameters corresponding to the request data based on the neural network model corresponding to multiple request parameter types to obtain more accurate request parameters, which is convenient for subsequent computing power resource scheduling, and assists in realizing more efficient and reasonable computing power scheduling for computing power requests of multiple users, thereby improving the efficiency and effectiveness of computing power resource scheduling and providing a better user experience for computing power leasing users.

[0040] As an optional embodiment, in the above steps, the neural network model is an RNN neural network model, which is trained by a training data set including request parameter annotations of multiple training request data and corresponding request parameter types, and optimizes model parameters based on a gradient descent algorithm and a cross entropy loss function until convergence.

[0041] It can be seen that through the above optional embodiments, the model details and training details of the neural network model are limited, which can predict more accurate request parameters, facilitate subsequent computing power resource scheduling, and assist in more efficient and reasonable computing power scheduling for computing power requests of multiple users, thereby improving the efficiency and effectiveness of computing power resource scheduling and providing a better user experience for computing power leasing users.

[0042] As an optional embodiment, in the above steps, the predicted computing power indicator parameter is the computing power task energy consumption requirement, the computing power task time limit, the computing power task performance requirement or the computing power device type requirement.

[0043] It can be seen that through the above optional embodiments, the types of predicted computing power indicator parameters are clarified, which can comprehensively characterize the user's computing power demand indicators, facilitate subsequent computing power resource scheduling, and assist in more efficient and reasonable computing power scheduling for multiple users' computing power requests, thereby improving the efficiency and effectiveness of computing power resource scheduling and providing a better user experience for computing power leasing users.

[0044] As an optional embodiment, in the above steps, based on the objective function and constraint conditions associated with the multiple predicted computing power indicator parameters, the computing power resource scheduling strategies corresponding to the multiple computing power requests are calculated according to the dynamic programming algorithm, including: Screening out multiple target computing devices that meet the computing device type requirements from multiple candidate computing devices; Obtain the device parameters corresponding to each target computing power device; optionally, the device parameters include device computing performance, device energy consumption, and device working stability; Determine the objective function and constraints based on the energy consumption requirements, time limits, and performance requirements of the computing tasks, as well as the device parameters corresponding to each target computing device; Based on the objective function and constraints, the computing power scheduling schemes corresponding to all target computing power devices and all computing power requests are iteratively calculated according to the dynamic programming algorithm until the optimal computing power scheduling scheme is obtained as the computing power resource scheduling strategy corresponding to multiple computing power requests.

[0045] It can be seen that through the above optional embodiments, multiple target computing power devices can be screened out according to the computing power device type requirements, and the optimal computing power scheduling plan can be calculated based on the determined objective function and constraints according to the dynamic programming algorithm, so as to achieve more efficient and reasonable computing power scheduling for the computing power requests of multiple users, improve the efficiency and effect of computing power resource scheduling, and provide a better user experience for computing power leasing users.

[0046] As an optional embodiment, in the above steps, according to the energy consumption requirement of the computing task, the time limit of the computing task and the performance requirement of the computing task, and the device parameters corresponding to each target computing device, the objective function and the constraint conditions are determined, including: Determine the objective function so that the number of computing power requests allocated to each target computing power device is minimized; Identify constraints including: The ratio between the first product value of the number of all computing power requests allocated to each target computing power device and the computing power task energy consumption requirement and the device energy consumption corresponding to the target computing power device is less than the first ratio threshold; The ratio between the second product value of the number of all computing power requests allocated to each target computing power device and the computing power task time limit and the working stability corresponding to the target computing power device is less than the second ratio threshold; The ratio between the third product value of the number of all computing power requests allocated to each target computing power device and the computing power task performance requirement and the device computing performance corresponding to the target computing power device is less than the third ratio threshold.

[0047] It can be seen that through the above optional embodiments, the objective function and constraints can be determined according to parameters such as device parameters, computing power task energy consumption requirements, computing power task time limits and computing power task performance requirements, so as to calculate the optimal computing power scheduling plan according to the dynamic programming algorithm, and realize more efficient and reasonable computing power scheduling for computing power requests of multiple users, improve the efficiency and effectiveness of computing power resource scheduling, and provide better user experience for computing power leasing users.

[0048] Embodiment 2 See also Figure 2 , Figure 2 : is a schematic diagram of a computing power scheduling system based on multi-model collaboration disclosed in an embodiment of the present invention. Figure 2 The computing power scheduling system based on multi-model collaboration described above can be applied to a data processing system / data processing device / data processing server (wherein the server includes a local processing server or a cloud processing server). Figure 2 As shown, the computing power scheduling system based on multi-model collaboration may include: The acquisition module 201 is used to acquire multiple computing power requests from multiple computing power users.

[0049] The determination module 202 is used to determine the request parameters corresponding to each computing power request according to the request data of each computing power request. The prediction module 203 is used to input the request parameters corresponding to all computing power requests into the prediction models corresponding to multiple computing power indicator types, so as to obtain corresponding multiple predicted computing power indicator parameters. The scheduling module 204 is used to calculate the computing power resource scheduling strategy corresponding to the multiple computing power requests based on the objective function and constraint conditions associated with the multiple predicted computing power indicator parameters according to the dynamic programming algorithm.

[0050] It can be seen that the above-mentioned embodiments of the invention can determine the corresponding request parameters according to the request data of each computing power request, and then predict the corresponding multiple predicted computing power indicator parameters based on the prediction models corresponding to the multiple computing power indicator types, so as to calculate a reasonable and efficient computing power resource scheduling strategy based on the objective function and constraint conditions associated with the predicted computing power indicator parameters, thereby being able to more efficiently and reasonably implement computing power scheduling for the computing power requests of multiple users, improve the efficiency and effect of computing power resource scheduling, and provide a better user experience for computing power leasing users.

[0051] As an optional embodiment, the request data includes computing power user information, request data, request device information, request network environment information and request location information.

[0052] It can be seen that through the above optional embodiments, the content of the request data is clarified, and the data content of the user's computing power request can be comprehensively characterized to facilitate subsequent computing power resource scheduling, assist in achieving more efficient and reasonable computing power scheduling for multiple users' computing power requests, improve the efficiency and effectiveness of computing power resource scheduling, and provide a better user experience for computing power leasing users.

[0053] As an optional embodiment, the request parameters include the requested task type, the requested task urgency, the requested personnel level, the requested task industry category and the requested malicious possibility parameter.

[0054] It can be seen that through the above optional embodiments, the content of the request parameters is clarified, which can comprehensively characterize the characteristics of the user's computing power request, so as to facilitate the subsequent computing power resource scheduling, and assist in realizing more efficient and reasonable computing power scheduling for multiple users' computing power requests, thereby improving the efficiency and effectiveness of computing power resource scheduling, and providing a better user experience for computing power leasing users.

[0055] As an optional embodiment, the determination module determines the specific manner of the request parameters corresponding to each computing power request according to the request data of each computing power request, including: For each computing power request, obtain the request data corresponding to the computing power request; The request data is input into a neural network model corresponding to multiple request parameter types to obtain the request parameters corresponding to the computing power request; the request parameter types are the request task type, the request task urgency, the request personnel level, the request task industry category or the request malicious possibility parameter.

[0056] It can be seen that through the above-mentioned optional embodiments, it is possible to predict the request parameters corresponding to the request data based on the neural network model corresponding to multiple request parameter types to obtain more accurate request parameters, which is convenient for subsequent computing power resource scheduling, and assists in realizing more efficient and reasonable computing power scheduling for computing power requests of multiple users, thereby improving the efficiency and effectiveness of computing power resource scheduling and providing a better user experience for computing power leasing users.

[0057] As an optional embodiment, the neural network model is an RNN neural network model, which is trained by a training data set including request parameter annotations of multiple training request data and corresponding request parameter types, and optimizes model parameters based on a gradient descent algorithm and a cross entropy loss function until convergence.

[0058] It can be seen that through the above optional embodiments, the model details and training details of the neural network model are limited, which can predict more accurate request parameters, facilitate subsequent computing power resource scheduling, and assist in more efficient and reasonable computing power scheduling for computing power requests of multiple users, thereby improving the efficiency and effectiveness of computing power resource scheduling and providing a better user experience for computing power leasing users.

[0059] As an optional embodiment, the predicted computing power indicator parameter is the computing power task energy consumption requirement, the computing power task time limit, the computing power task performance requirement or the computing power device type requirement.

[0060] It can be seen that through the above optional embodiments, the types of predicted computing power indicator parameters are clarified, which can comprehensively characterize the user's computing power demand indicators, facilitate subsequent computing power resource scheduling, and assist in more efficient and reasonable computing power scheduling for multiple users' computing power requests, thereby improving the efficiency and effectiveness of computing power resource scheduling and providing a better user experience for computing power leasing users.

[0061] As an optional embodiment, the scheduling module calculates the specific manner of the computing resource scheduling strategy corresponding to the multiple computing power requests based on the objective function and constraint conditions associated with the multiple predicted computing power indicator parameters according to the dynamic programming algorithm, including: Screening out multiple target computing devices that meet the computing device type requirements from multiple candidate computing devices; Obtain the device parameters corresponding to each target computing power device; optionally, the device parameters include device computing performance, device energy consumption, and device working stability; Determine the objective function and constraints based on the energy consumption requirements, time limits, and performance requirements of the computing tasks, as well as the device parameters corresponding to each target computing device; Based on the objective function and constraints, the computing power scheduling schemes corresponding to all target computing power devices and all computing power requests are iteratively calculated according to the dynamic programming algorithm until the optimal computing power scheduling scheme is obtained as the computing power resource scheduling strategy corresponding to multiple computing power requests.

[0062] It can be seen that through the above optional embodiments, multiple target computing power devices can be screened out according to the computing power device type requirements, and the optimal computing power scheduling plan can be calculated based on the determined objective function and constraints according to the dynamic programming algorithm, so as to achieve more efficient and reasonable computing power scheduling for the computing power requests of multiple users, improve the efficiency and effect of computing power resource scheduling, and provide a better user experience for computing power leasing users.

[0063] As an optional embodiment, the scheduling module determines the specific manner of the objective function and the constraint conditions according to the energy consumption requirements of the computing task, the time limit of the computing task, the performance requirements of the computing task, and the device parameters corresponding to each target computing device, including: Determine the objective function so that the number of computing power requests allocated to each target computing power device is minimized; Identify constraints including: The ratio between the first product value of the number of all computing power requests allocated to each target computing power device and the computing power task energy consumption requirement and the device energy consumption corresponding to the target computing power device is less than the first ratio threshold; The ratio between the second product value of the number of all computing power requests allocated to each target computing power device and the computing power task time limit and the working stability corresponding to the target computing power device is less than the second ratio threshold; The ratio between the third product value of the number of all computing power requests allocated to each target computing power device and the computing power task performance requirement and the device computing performance corresponding to the target computing power device is less than the third ratio threshold.

[0064] It can be seen that through the above optional embodiments, the objective function and constraints can be determined according to parameters such as device parameters, computing power task energy consumption requirements, computing power task time limits and computing power task performance requirements, so as to calculate the optimal computing power scheduling plan according to the dynamic programming algorithm, and realize more efficient and reasonable computing power scheduling for computing power requests of multiple users, improve the efficiency and effectiveness of computing power resource scheduling, and provide better user experience for computing power leasing users.

[0065] Embodiment 3 See also Figure 3 , Figure 3 It is another computing power scheduling system based on multi-model collaboration disclosed in an embodiment of the present invention. Figure 3 The computing power scheduling system based on multi-model collaboration is applied to a data processing system / data processing device / data processing server (wherein the server includes a local processing server or a cloud processing server). Figure 3 As shown, the computing power scheduling system based on multi-model collaboration may include: A memory 301 storing executable program codes; a processor 302 coupled to the memory 301; Among them, the processor 302 calls the executable program code stored in the memory 301 to execute the steps of the computing power scheduling method based on multi-model collaboration described in Example 1.

[0066] Embodiment 4 An embodiment of the present invention discloses a computer-readable storage medium that stores a computer program for electronic data exchange, wherein the computer program enables a computer to execute the steps of the computing power scheduling method based on multi-model collaboration described in Embodiment 1.

[0067] Embodiment 5 An embodiment of the present invention discloses a computer program product, which includes a non-transitory computer-readable storage medium storing a computer program, and the computer program is operable to enable a computer to execute the steps of the computing power scheduling method based on multi-model collaboration described in Example 1.

[0068] The above describes specific embodiments of the present specification, and other embodiments are within the scope of the appended claims. In some cases, the actions or steps recorded in the claims can be performed in an order different from that in the embodiments and still achieve the desired results. In addition, the processes depicted in the accompanying drawings do not necessarily have to be performed in the specific order or sequential order shown to achieve the desired results. In some embodiments, multitasking and parallel processing are also possible or may be advantageous.

[0069] The systems, devices, modules or units described in the above embodiments may be implemented by computer chips or entities, or by products with certain functions. A typical implementation device is a computer. Specifically, the computer may be, for example, a personal computer, a laptop computer, a cellular phone, a camera phone, a smart phone, a personal digital assistant, a media player, a navigation device, an email device, a game console, a tablet computer, a wearable device, or a combination of any of these devices.

[0070] For the convenience of description, the above device is described by dividing it into various units according to its functions. Of course, when implementing this specification, the functions of each unit can be implemented in the same or multiple software and / or hardware.

[0071] Those skilled in the art will appreciate that the embodiments of this specification may be provided as methods, systems, or computer program products. Therefore, the embodiments of this specification may be in the form of complete hardware embodiments, complete software embodiments, or embodiments in combination with software and hardware. Moreover, the embodiments of this specification may be in the form of a computer program product implemented in one or more computer-usable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) that contain computer-usable program code.

[0072] This specification is described with reference to the flowcharts and / or block diagrams of the methods, devices (systems), and computer program products according to the embodiments of this specification. It should be understood that each process and / or box in the flowchart and / or block diagram, as well as the combination of the processes and / or boxes in the flowchart and / or block diagram, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, a special-purpose computer, an embedded processor, or other programmable data processing device to produce a machine, so that the instructions executed by the processor of the computer or other programmable data processing device generate instructions for implementing the processes in the flowchart and / or block diagram. Figure 1 A process or multiple processes and / or boxes Figure 1 A device that provides the functions specified in a block or multiple blocks.

[0073] These computer program instructions may also be stored in a computer-readable memory capable of directing a computer or other programmable data processing device to operate in a specific manner, so that the instructions stored in the computer-readable memory produce an article of manufacture comprising an instruction device, which implements the process Figure 1 A process or multiple processes and / or boxes Figure 1 A function specified in one or more boxes.

[0074] These computer program instructions can also be loaded onto a computer or other programmable data processing device so that a series of operating steps are executed on the computer or other programmable device to produce a computer-implemented process, thereby providing instructions for implementing the process. Figure 1 A process or multiple processes and / or boxes Figure 1 A step that specifies a function in one or more boxes.

[0075] In a typical configuration, a computing device includes one or more processors (CPU), input / output interfaces, network interfaces, and memory.

[0076] The memory may include non-permanent storage in a computer-readable medium, random access memory (RAM) and / or non-volatile memory in the form of read-only memory (ROM) or flash RAM. The memory is an example of a computer-readable medium.

[0077] Computer readable media include permanent and non-permanent, removable and non-removable media that can be implemented by any method or technology to store information. Information can be computer readable instructions, data structures, program modules or other data. Examples of computer storage media include, but are not limited to, phase change memory (PRAM), static random access memory (SRAM), dynamic random access memory (DRAM), other types of random access memory (RAM), read-only memory (ROM), electrically erasable programmable read-only memory (EEPROM), flash memory or other memory technology, compact disk read-only memory (CD-ROM), digital versatile disk (DVD) or other optical storage, magnetic cassettes, magnetic disk storage or other magnetic storage devices or any other non-transmission media that can be used to store information that can be accessed by a computing device. As defined herein, computer readable media does not include temporary computer readable media (transitory media), such as modulated data signals and carrier waves.

[0078] It should also be noted that the terms "include", "comprises" or any other variations thereof are intended to cover non-exclusive inclusion, so that a process, method, commodity or device including a series of elements includes not only those elements, but also other elements not explicitly listed, or also includes elements inherent to such process, method, commodity or device. In the absence of more restrictions, the elements defined by the sentence "comprises a ..." do not exclude the existence of other identical elements in the process, method, commodity or device including the elements.

[0079] This specification may be described in the general context of computer-executable instructions executed by a computer, such as program modules. Generally, program modules include routines, programs, objects, components, data structures, etc. that perform specific tasks or implement specific abstract data types. This specification may also be practiced in distributed computing environments where tasks are performed by remote processing devices connected through a communication network. In a distributed computing environment, program modules may be located in local and remote computer storage media, including storage devices.

[0080] Each embodiment in this specification is described in a progressive manner, and the same or similar parts between the embodiments can be referred to each other, and each embodiment focuses on the differences from other embodiments. In particular, for the system embodiment, since it is basically similar to the method embodiment, the description is relatively simple, and the relevant parts can be referred to the partial description of the method embodiment.

[0081] Finally, it should be noted that the computing power scheduling method and system based on multi-model collaboration disclosed in the embodiments of the present invention only discloses the preferred embodiments of the present invention, which are only used to illustrate the technical scheme of the present invention, rather than to limit it. Although the present invention has been described in detail with reference to the aforementioned embodiments, those skilled in the art should understand that the technical schemes described in the aforementioned embodiments can still be modified, or some of the technical features can be replaced by equivalents. However, these modifications or replacements do not deviate the essence of the corresponding technical schemes from the spirit and scope of the technical schemes of the embodiments of the present invention.

Claims

1. A computing power scheduling method based on multi-model collaboration, characterized in that: The method comprises: Get multiple computing power requests from multiple computing power users; Determine, according to the request data of each computing power request, a request parameter corresponding to each computing power request; Inputting the request parameters corresponding to all the computing power requests into the prediction models corresponding to the multiple computing power indicator types respectively to obtain the corresponding multiple predicted computing power indicator parameters; Based on the objective functions and constraints associated with the multiple predicted computing power indicator parameters, a computing power resource scheduling strategy corresponding to the multiple computing power requests is calculated according to a dynamic programming algorithm.

2. The computing power scheduling method based on multi-model collaboration according to claim 1 is characterized in that: The request data includes computing power user information, request data, request device information, request network environment information and request location information.

3. The computing power scheduling method based on multi-model collaboration according to claim 1 is characterized in that: The request parameters include the requested task type, the requested task urgency, the requested personnel level, the requested task industry category and the requested malicious possibility parameter.

4. The computing power scheduling method based on multi-model collaboration according to claim 3 is characterized in that: The determining, according to the request data of each computing power request, a request parameter corresponding to each computing power request includes: For each computing power request, obtaining request data corresponding to the computing power request; The request data is input into a neural network model corresponding to a plurality of request parameter types to obtain request parameters corresponding to the computing power request; the request parameter types are request task type, request task urgency, request personnel level, request task industry category or request malicious possibility parameter.

5. The computing power scheduling method based on multi-model collaboration according to claim 4 is characterized in that: The neural network model is an RNN neural network model, which is trained by a training data set including a plurality of training request data and request parameter annotations of corresponding request parameter types, and optimizes model parameters based on a gradient descent algorithm and a cross entropy loss function until convergence.

6. The computing power scheduling method based on multi-model collaboration according to claim 1 is characterized in that: The predicted computing power indicator parameter is computing power task energy consumption requirement, computing power task time limit, computing power task performance requirement or computing power device type requirement.

7. The computing power scheduling method based on multi-model collaboration according to claim 6 is characterized in that: The step of calculating the computing resource scheduling strategies corresponding to the multiple computing power requests based on the objective functions and constraints associated with the multiple predicted computing power indicator parameters according to the dynamic programming algorithm includes: Screening out multiple target computing devices that meet the computing device type requirements from multiple candidate computing devices; Obtaining device parameters corresponding to each of the target computing power devices; the device parameters include device computing performance, device energy consumption, and device working stability; Determine the objective function and constraint conditions according to the energy consumption requirement of the computing task, the time limit of the computing task, the performance requirement of the computing task, and the device parameters corresponding to each of the target computing devices; Based on the objective function and the constraints, the computing power scheduling schemes corresponding to all the target computing power devices and all the computing power requests are iteratively calculated according to the dynamic programming algorithm until the optimal computing power scheduling scheme is obtained as the computing power resource scheduling strategy corresponding to the multiple computing power requests.

8. The computing power scheduling method based on multi-model collaboration according to claim 7 is characterized in that: Determining the objective function and the constraint conditions according to the energy consumption requirement of the computing task, the time limit of the computing task, the performance requirement of the computing task, and the device parameters corresponding to each of the target computing devices includes: Determine the objective function so that the number of computing power requests allocated to each of the target computing power devices is minimized; Identify constraints including: The ratio between the first product value of the number of all computing power requests allocated to each of the target computing power devices and the computing power task energy consumption requirement and the device energy consumption corresponding to the target computing power device is less than a first ratio threshold; The ratio between the number of all computing power requests allocated to each of the target computing power devices and the second product value of the computing power task time limit and the working stability corresponding to the target computing power device is less than a second ratio threshold; The ratio between the third product value of the number of all computing power requests allocated to each of the target computing power devices and the computing power task performance requirement and the device computing performance corresponding to the target computing power device is less than a third ratio threshold.

9. A computing power scheduling system based on multi-model collaboration, characterized in that: The system comprises: An acquisition module, used to acquire multiple computing power requests from multiple computing power users; A determination module, used to determine the request parameters corresponding to each computing power request according to the request data of each computing power request; A prediction module, used to input the request parameters corresponding to all the computing power requests into the prediction models corresponding to the multiple computing power indicator types, so as to obtain the corresponding multiple predicted computing power indicator parameters; The scheduling module is used to calculate the computing power resource scheduling strategy corresponding to the multiple computing power requests based on the objective function and constraints associated with the multiple predicted computing power indicator parameters and according to the dynamic programming algorithm.

10. A computing power scheduling system based on multi-model collaboration, characterized in that: The system comprises: A memory storing executable program code; a processor coupled to the memory; The processor calls the executable program code stored in the memory to execute the computing power scheduling method based on multi-model collaboration as described in any one of claims 1-8.

Images