Computing power resource management and distribution method and system and medium

Abstract

The invention provides a computing power resource management and distribution method and system and a medium. The method comprises the following steps: integrating various computing power resources in a private network, establishing a computing power resource information base, analyzing a computing power resource request, establishing a service demand model, optimizing and matching a computing power resource configuration template, obtaining a computing power adaptation template, and establishing a service demand model; optimal computing power resource adaptation is carried out according to dynamic information of target computing power resources matched with the computing power adaptation template, isolation processing is carried out at the same time, monitored performance index data are processed, a load trend is predicted, a corresponding computing power resource adjustment strategy is obtained, and the computing power resources are optimized and expanded according to big data analysis; therefore, efficient utilization and elastic expansion of the computing power resources of the private network are realized by constructing the dynamic computing power resource pool and the resource scheduling model and monitoring, adjusting and optimizing an expansion mechanism, and the processing efficiency and security of big data services are remarkably improved.

Description

Technical Field

[0001] This application relates to the technical field of big data processing, and more particularly, to a computing power resource management and allocation method, system, and medium. Background Art

[0002] With the construction and development of the national big data system, the demand for computing power resources in big data business systems is increasing day by day. There are problems such as unreasonable resource allocation and inability to dynamically adapt to changes in business load in the big data private network, resulting in waste of computing power resources or limited performance of business systems, and it is difficult to fully ensure the efficient operation of big data services.

[0003] In view of the above problems, there is an urgent need for effective technical solutions. Summary of the Invention

[0004] The purpose of this application is to provide a computing power resource management and allocation method, system, and medium. By constructing a dynamic computing power resource pool, a resource scheduling model, monitoring, adjustment, and optimization expansion, the efficient utilization and elastic expansion of computing power resources in the private network are realized, and the processing efficiency and security of big data services are significantly improved.

[0005] This application also provides a computing power resource management and allocation method, including the following steps: Integrate various types of computing power resources in a predetermined private network to establish a computing power resource information database; Analyze the computing power resource requests of the service, establish a service demand model, optimize and match the computing power resource configuration template, and obtain a computing power adaptation template; Perform optimal computing power resource adaptation according to the dynamic information of the target computing power resources matched by the computing power adaptation template, and at the same time perform isolation processing; Process the performance index data of the monitored optimal computing power resources, predict the load trend, and obtain the corresponding computing power resource adjustment strategy; Process the corresponding computing power resource historical data combined with the load trend data to obtain a computing power resource expansion plan, and expand and deploy the service computing power resources.

[0006] Optionally, in a computing power resource management and allocation method described in this application, the integrating various types of computing power resources in a predetermined private network to establish a computing power resource information database includes: Integrate various types of computing power resource modules in a predetermined private network to form a computing power resource pool; Perform performance evaluation and feature description on each preset computing power unit in the computing power resource pool to establish a computing power resource information database.

[0007] Optionally, in a computing power resource management and allocation method described in this application, the analysis of the computing power resource requests for services, the establishment of a service demand model, and the optimization and matching of the computing power resource configuration template to obtain a computing power adaptation template include: Analyze the computing power resource requests for services, and extract service feature information, including service type, load characteristics, performance requirements, and service attributes; Establish a service demand model based on the service type, load characteristics, and performance requirements; Obtain multiple matching computing power resource configuration templates from the computing power resource information library according to the service attributes in combination with the preset service priorities; Optimize and match the multiple computing power resource configuration templates according to the service demand model to obtain a computing power adaptation template.

[0008] Optionally, in a computing power resource management and allocation method described in this application, the optimal computing power resource adaptation based on the dynamic information of the target computing power resources matched by the computing power adaptation template, and at the same time, isolation processing is performed, including: Match the target computing power resources in the computing power resource pool according to the computing power adaptation template; Obtain the dynamic information of the target computing power resources, including deployment location, load status, and compatibility; Perform optimal computing power resource adaptation for the service according to the dynamic information through a preset resource scheduling model; Perform isolation processing on the computing power resource adaptation through a preset containerization model and a software-defined network model.

[0009] Optionally, in a computing power resource management and allocation method described in this application, the processing of the performance index data of the monitored optimal computing power resources, predicting the load trend, and obtaining the corresponding computing power resource adjustment strategy include: Monitor the running status of the adapted optimal computing power resources in real time, and extract the performance index data; The performance index data includes CPU utilization rate, memory occupancy rate, disk read and write speed, and network traffic; Process the performance index data through a preset load prediction model to obtain load trend data; Compare the load trend data with a preset load prediction threshold to obtain the corresponding computing power resource adjustment strategy.

[0010] Optionally, in a computing power resource management and allocation method described in this application, the processing of the corresponding computing power resource historical data in combination with the load trend data to obtain a computing power resource expansion plan, and the expansion deployment of the service computing power resources includes: Obtain the historical data of computing power resources for the optimal computing power adaptation template of the service, including the average utilization rate of computing power resources and the coordination effective coefficient of computing power resources; Process the average utilization rate of computing power resources and the coordination effective coefficient of computing power resources in combination with the load trend data to obtain the computing power optimization degree coefficient; Query through the preset computing power optimization configuration queue list according to the computing power optimization degree coefficient to obtain a computing power resource expansion plan, including hardware upgrade or policy optimization improvement; Expand and deploy the service computing power resources according to the computing power resource expansion plan.

[0011] In a second aspect, the present application provides a computing power resource management and allocation system, including: Computing power resource management module: Integrate various computing power resource modules in a predetermined private network to form a computing power resource pool, evaluate the performance of each preset computing power unit therein, and establish a computing power resource information database; Business requirement analysis module: Analyze the computing power resource requests of the service, establish a business requirement model and optimize and match the computing power resource configuration template to obtain a computing power adaptation template and obtain computing power resource allocation suggestions; Computing power resource scheduling module: Perform optimal computing power resource adaptation on the dynamic information of the target computing power resources matched by the computing power adaptation template and perform isolation processing at the same time; Monitoring and adjustment module: Monitor the performance index data of the optimal computing power resources and predict the load trend to obtain the corresponding computing power resource adjustment strategy; Optimization and expansion module: Conduct a computing power optimization degree evaluation based on the historical data of computing power resources corresponding to the optimal computing power adaptation template in combination with the load trend data to obtain the corresponding computing power resource expansion plan, and expand and deploy the service computing power resources.

[0012] Optionally, in a computing power resource management and allocation system described in the present application, the system includes: a memory and a processor. A program for a computing power resource management and allocation method is included in the memory. When the program for the computing power resource management and allocation method is executed by the processor, the following steps are implemented: Integrate various computing power resources in a predetermined private network and establish a computing power resource information database; Analyze the computing power resource requests of the service, establish a business requirement model and optimize and match the computing power resource configuration template to obtain a computing power adaptation template; Perform optimal computing power resource adaptation on the dynamic information of the target computing power resources matched by the computing power adaptation template and perform isolation processing at the same time; Process according to the monitored performance index data of the optimal computing power resources, predict the load trend and obtain the corresponding computing power resource adjustment strategy; Based on the corresponding historical data of computing power resources and combined with the load trend data, a computing power resource expansion plan is obtained through processing, and the business computing power resources are expanded and deployed.

[0013] Optionally, in a computing power resource management and allocation system described in this application, the integration of various types of computing power resources in a predetermined private network to establish a computing power resource information library includes: Integrate various types of computing power resource modules in a predetermined private network to form a computing power resource pool; Perform performance evaluation and feature description on each preset computing unit in the computing power resource pool to establish a computing power resource information library.

[0014] In a third aspect, this application also provides a computer-readable storage medium. A computing power resource management and allocation method program is stored in the computer-readable storage medium. When the computing power resource management and allocation method program is executed by a processor, the steps of a computing power resource management and allocation method as described in any one of the above are implemented.

[0015] As can be seen from the above, a computing power resource management and allocation method, system, and medium provided by this application integrate various types of computing power resources in a private network to establish a computing power resource information library, analyze computing power resource requests, establish a business demand model and optimize and match the computing power resource configuration template to obtain a computing power adaptation template, perform optimal computing power resource adaptation according to the dynamic information of the target computing power resources matched by the computing power adaptation template, and at the same time perform isolation processing, process the monitored performance index data, predict the load trend and obtain the corresponding computing power resource adjustment strategy, and optimize and expand the computing power resources according to big data analysis; thereby, through constructing a dynamic computing power resource pool, a resource scheduling model, monitoring, adjustment, optimization, and expansion, the efficient utilization and elastic expansion of the computing power resources of the private network are realized, and the processing efficiency and security of big data services are significantly improved.

[0016] Other features and advantages of this application will be described in the subsequent description. Moreover, some of them will become obvious from the description or can be understood by implementing the embodiments of this application. The objectives and other advantages of this application can be achieved and obtained through the structures specifically pointed out in the written description and the drawings. Description of the Drawings

[0017] To more clearly illustrate the technical solutions of the embodiments of this application, the following will briefly introduce the drawings required to be used in the embodiments of this application. It should be understood that the following drawings only show some embodiments of this application, and thus should not be regarded as limiting the scope. For those of ordinary skill in the art, other related drawings can be obtained based on these drawings without creative efforts.

[0018] Figure 1Flowchart of a computing power resource management and allocation method provided by an embodiment of the present application; Figure 2 Flowchart of constructing a computing power resource pool for a computing power resource management and allocation method provided by an embodiment of the present application; Figure 3 Flowchart of business requirement analysis and modeling for a computing power resource management and allocation method provided by an embodiment of the present application; Figure 4 Flowchart of computing power resource allocation and scheduling for a computing power resource management and allocation method provided by an embodiment of the present application; Figure 5 System diagram of a computing power resource management and allocation system provided by an embodiment of the present application. Detailed implementation manners

[0019] Next, the technical solutions in the embodiments of the present application will be clearly and completely described in conjunction with the accompanying drawings in the embodiments of the present application. Obviously, the described embodiments are only a part of the embodiments of the present application, rather than all the embodiments. The components of the embodiments of the present application described and illustrated herein can be arranged and designed in various different configurations. Therefore, the detailed description of the embodiments of the present application provided in the accompanying drawings is not intended to limit the scope of the present application to be protected, but only represents the selected embodiments of the present application. All other embodiments obtained by those skilled in the art based on the embodiments of the present application without creative efforts fall within the scope of protection of the present application.

[0020] It should be noted that similar reference numerals and letters denote similar items in the following drawings. Therefore, once an item is defined in one drawing, it does not need to be further defined and explained in subsequent drawings. At the same time, in the description of the present application, the terms "first", "second", etc. are only used for descriptive distinction and cannot be understood as indicating or implying relative importance.

[0021] Please refer to Figure 1 , Figure 1 is a flowchart of a computing power resource management and allocation method in some embodiments of the present application. This computing power resource management and allocation method is used in terminal devices, such as computers, mobile phone terminals, etc. This computing power resource management and allocation method includes the following steps: S11. Integrate various types of computing power resources in a predetermined private network to establish a computing power resource information database; S12. Analyze the computing power resource requests of the service, establish a business requirement model, and optimize and match the computing power resource configuration template to obtain a computing power adaptation template; S13. Perform optimal computing power resource adaptation according to the dynamic information of the target computing power resources matched by the computing power adaptation template, and perform isolation processing at the same time; S14. Process according to the performance index data of the monitored optimal computing power resources, predict the load trend and obtain the corresponding computing power resource adjustment strategy; S15. Process according to the corresponding historical data of computing power resources combined with the load trend data to obtain a computing power resource expansion plan, and deploy the expansion of business computing power resources.

[0022] Among them, various computing power resources in the predetermined private network are integrated to build a dynamic computing power resource pool. At the same time, through business demand analysis and modeling, and resource scheduling models, the processing efficiency and security of big data services are improved. Through the monitoring adjustment and optimization expansion mechanism, the efficient utilization and elastic expansion of computing power resources are realized, continuously meeting the requirements of business development.

[0023] Please refer to Figure 2 , Figure 2 is a flowchart of the construction of a computing power resource pool for a computing power resource management and allocation method in some embodiments of the present application. According to an embodiment of the present invention, the integration of various computing power resources in the predetermined private network and the establishment of a computing power resource information library include: S21. Integrate various computing power resource modules in the predetermined private network to form a computing power resource pool; S22. Perform performance evaluation and feature description on each preset computing power unit in the computing power resource pool, and establish a computing power resource information library.

[0024] Among them, the predetermined private network in this solution is adapted to the national cultural private network, and various computing power resources therein, including data center servers, cloud platforms, and edge computing devices, are integrated. Each preset computing power unit, such as a server and a GPU cluster, is subjected to performance testing and feature recording, including recording information such as CPU model, number of cores, capacity, storage type, and bandwidth, as well as software environment information such as operating system type and framework, so as to establish a dynamically updated computing power resource information library for convenient subsequent query and allocation.

[0025] Please refer to Figure 3 , Figure 3 is a flowchart of business demand analysis and modeling for a computing power resource management and allocation method in some embodiments of the present application. According to an embodiment of the present invention, the analysis of the computing power resource request of the business, the establishment of a business demand model and the optimization of the matching of the computing power resource configuration template to obtain a computing power adaptation template include: S31. Analyze the computing power resource request of the business, and extract business feature information, including business type, load characteristics, performance requirements, and business attributes; S32. Establish a business demand model according to the business type, load characteristics, and performance requirements; S33. Obtain multiple matching computing power resource configuration templates from the computing power resource information library according to the service attribute in combination with the preset service priority; S34. Optimally match the multiple computing power resource configuration templates according to the service demand model to obtain a computing power adaptation template.

[0026] Among them, the analysis of the computing power resource request includes analyzing service types such as compute-intensive and bandwidth-intensive, load characteristics such as periodic and bursty, performance requirements such as latency-sensitive and high throughput. The service attributes include the collection, annotation, and deconstruction of cultural resource data or the production and distribution of cultural digital content. Then, quantify the key performance indicators of the computing power resources according to the service type, load characteristics, and performance requirements. Establish a service demand model by correlating the service requirements and key performance indicators of each computing node of the computing power resources. For example, for a simple data sorting requirement, time response, data volume size, and complexity can be used as key performance indicators to establish a service demand model related to the computing power of the required CPU. Then, match multiple configuration templates according to the service attribute and priority. For example, for a service with high priority for annotation, a high-performance GPU needs to be configured to accelerate the operation of the deep learning algorithm model to improve the annotation efficiency and accuracy. Then, through the resource management central platform of the computing power resource information library, automatically adjust and match the computing power resource adaptation template with the established service demand model.

[0027] Please refer to Figure 4 , Figure 4 is a flowchart of the computing power resource allocation and scheduling in a computing power resource management and allocation method according to some embodiments of the present application. According to an embodiment of the present invention, the optimal computing power resource adaptation is performed according to the dynamic information of the target computing power resources matched by the computing power adaptation template, and isolation processing is performed simultaneously, including: S41. Match the target computing power resources in the computing power resource pool according to the computing power adaptation template; S42. Obtain the dynamic information of the target computing power resources, including the deployment location, load status, and compatibility; S43. Perform optimal computing power resource adaptation for the service according to the dynamic information through a preset resource scheduling model; S44. Perform isolation processing on the computing power resource adaptation through a preset containerization model and software-defined network model.

[0028] Among them, through a preset resource scheduling model such as a genetic algorithm model or a simulated annealing algorithm model, the dynamic deployment location, load, compatibility, and service priority of the matched target computing power resources are optimized and adapted for business resources to generate an optimal adaptation plan for resource allocation. For example, the simulated annealing algorithm performs vector encoding and fitness function establishment on the dynamic deployment location, load, compatibility, and service priority of computing power resources, and then obtains the optimal solution through iterative calculation of function differences, which is the optimal deployment plan. At the same time, it also dynamically allocates appropriate computing power resources for the service through the API interface with the cloud computing platform. During the process of optimizing resource allocation, through a preset containerization model such as K8s and a software-defined network model such as SDN, isolation processing is performed on the business resource allocation to create an independent operating environment, ensuring that multiple services do not interfere with each other during parallel allocation, and improving the high utilization rate and security of resources.

[0029] According to an embodiment of the present invention, the processing of the performance index data of the monitored optimal computing power resources, predicting the load trend, and obtaining the corresponding computing power resource adjustment strategy includes: Real-time monitor the operating status of the adapted optimal computing power resources and extract performance index data; The performance index data includes CPU utilization rate, memory occupancy rate, disk read and write speed, and network traffic; Process the performance index data through a preset load prediction model to obtain load trend data; Compare the load trend data with a preset load prediction threshold to obtain the corresponding computing power resource adjustment strategy.

[0030] Among them, the preset load prediction model is an operation processing model that calculates and processes the performance index data of the real-time collected operating status based on a preset algorithm. The preset algorithm can adopt algorithms such as regression algorithm, decision tree or random forest algorithm, LSTM, etc. For example, when using the linear regression algorithm, a linear relationship between CPU utilization rate, memory occupancy rate, disk read and write speed, and network traffic and the load is established, a linear equation is established to fit historical data, the weight coefficients of each index data are determined, and the future load trend is linearly predicted according to the coefficients and index values. Then, through threshold comparison and interval fitting with the preset load prediction threshold, the interval corresponding to the calculated load trend data is obtained, and the strategy corresponding to the interval is the adjustment strategy for computing power resources. For example, the threshold comparison intervals in the embodiment of this solution are [0, 0.32), [0.32, 0.48), [0.48, 0.79), [0.79, 1.0], corresponding to the first, second, third, and fourth computing power resource adjustment strategies respectively. If the threshold comparison result falls into the third interval, the computing power resources are adjusted according to the third strategy. The adjustment strategies include increasing / decreasing the number of CPU cores, increasing / decreasing the memory capacity, or releasing idle computing power resources back to the computing power resource pool.

[0031] According to an embodiment of the present invention, obtaining a computing power resource expansion plan by processing the corresponding historical data of computing power resources in combination with load trend data, and expanding and deploying the business computing power resources, includes: Obtaining the historical data of computing power resources of the optimal computing power adaptation template of the service, including the average utilization rate of computing power resources and the coordination effectiveness coefficient of computing power resources; Processing according to the average utilization rate of computing power resources and the coordination effectiveness coefficient of computing power resources in combination with the load trend data to obtain a computing power optimization degree coefficient; Querying according to the computing power optimization degree coefficient through a preset computing power optimization configuration queue list to obtain a computing power resource expansion plan, including hardware upgrade or strategy optimization improvement; Expanding and deploying the business computing power resources according to the computing power resource expansion plan.

[0032] Among them, the average utilization rate of computing power resources is the total average utilization rate of resources of the computing power adaptation template within a preset historical period, and the coordination effectiveness coefficient of computing power resources is an evaluation coefficient (with a value range of 0-1) of the coordination effectiveness of the computing power resource management and control platform for the computing power adaptation template within the historical period. Then, in combination with the load trend data, weighted calculation is performed to obtain the computing power optimization degree coefficient, and then the expansion plan is obtained through list query. Among them, hardware upgrade includes upgrading the CPU, increasing memory, and upgrading storage devices, and strategy optimization improvement includes optimizing the resource allocation area, adjusting the allocation priority, and improving the load balancing algorithm, so as to improve the overall performance and efficiency of computing power resources and ensure continuous satisfaction of the computing power requirements of the business.

[0033] Please refer to Figure 5 , Figure 5 which is a system diagram of a computing power resource management and allocation system in some embodiments of the present application.

[0034] In a second aspect, the present invention also discloses a computing power resource management and allocation system 5, which includes: A computing power resource management module 501: integrating various computing power resource modules in a predetermined private network to form a computing power resource pool, evaluating the performance of each preset computing power unit therein, and establishing a computing power resource information database; A service demand analysis module 502: analyzing the computing power resource requests of the service, establishing a service demand model and optimizing and matching the computing power resource configuration template to obtain a computing power adaptation template and obtain computing power resource allocation suggestions; A computing power resource scheduling module 503: performing optimal computing power resource adaptation on the dynamic information of the target computing power resources matched by the computing power adaptation template, and simultaneously performing isolation processing; A monitoring and adjustment module 504: monitoring the performance index data obtained from the optimal computing power resources and predicting the load trend to obtain the corresponding computing power resource adjustment strategy; Optimization and Expansion Module 505: Based on the historical data of computing power resources corresponding to the optimal computing power adaptation template and combined with the load trend data, conduct an evaluation of the computing power optimization degree, obtain the corresponding computing power resource expansion plan, and expand and deploy the business computing power resources.

[0035] According to an embodiment of the present invention, the computing power resource management and allocation system further includes: a memory and a processor. The memory includes the program of the computing power resource management and allocation method. When the program of the computing power resource management and allocation method is executed by the processor, the following steps are implemented: Integrate various types of computing power resources in a predetermined private network to establish a computing power resource information library; Analyze the computing power resource requests of the business, establish a business demand model, optimize and match the computing power resource configuration template, and obtain a computing power adaptation template; Perform optimal computing power resource adaptation according to the dynamic information of the target computing power resources matched by the computing power adaptation template, and at the same time perform isolation processing; Process according to the monitored performance index data of the optimal computing power resources, predict the load trend, and obtain the corresponding computing power resource adjustment strategy; Process according to the corresponding historical data of computing power resources combined with the load trend data to obtain a computing power resource expansion plan, and expand and deploy the business computing power resources.

[0036] Among them, integrate various types of computing power resources in the predetermined private network to build a dynamic computing power resource pool. At the same time, through business demand analysis and modeling, and resource scheduling models, improve the processing efficiency and security of big data services. Through monitoring, adjustment and optimization and expansion mechanisms, achieve efficient utilization and elastic expansion of computing power resources, and continuously meet the requirements of business development.

[0037] According to an embodiment of the present invention, the integration of various types of computing power resources in the predetermined private network to establish a computing power resource information library includes: Integrate various types of computing power resource modules in the predetermined private network to form a computing power resource pool; Perform performance evaluation and feature description on each preset computing power unit in the computing power resource pool to establish a computing power resource information library.

[0038] Among them, the predetermined private network in this solution is adapted to the national cultural private network. Integrate various types of computing power resources therein, including data center servers, cloud platforms, and edge computing devices. For each preset computing power unit, such as servers and GPU clusters, perform performance tests and feature records on each computing power unit, including recording information such as CPU model, number of cores, capacity, storage type and bandwidth, as well as software environment information such as operating system type and framework, so as to establish a dynamically updated computing power resource information library for convenient subsequent query and allocation.

[0039] According to an embodiment of the present invention, the analysis of the computing power resource request for a service, establishing a service demand model and optimizing and matching a computing power resource configuration template to obtain a computing power adaptation template includes: Analyze the computing power resource request for the service, and extract service feature information, including service type, load characteristics, performance requirements, and service attributes; Establish a service demand model according to the service type, load characteristics, and performance requirements; Obtain multiple matching computing power resource configuration templates through the computing power resource information library according to the service attributes in combination with a preset service priority; Optimize and match the multiple computing power resource configuration templates according to the service demand model to obtain a computing power adaptation template.

[0040] Among them, the analysis of the computing power resource request includes analyzing the service type such as compute-intensive and bandwidth-intensive, load characteristics such as periodic and bursty, performance requirements such as latency-sensitive and high-throughput, and service attributes including the collection, annotation, and deconstruction of cultural resource data or the production and distribution of cultural digital content. Then, quantify the key performance indicators of the computing power resources according to the service type, load characteristics, and performance requirements, and establish a service demand model by correlating the service requirements and key performance indicators of each computing node of the computing power resources. For example, for a simple data sorting requirement, time response, data volume size, and complexity can be used as key performance indicators to establish a service demand model related to the computing power of the required CPU. Then, match multiple configuration templates according to the service attributes and priorities. For example, for a high-priority annotation service, a high-performance GPU needs to be configured to accelerate the operation of the deep learning algorithm model to improve the annotation efficiency and accuracy. Finally, through the resource management central platform of the computing power resource information library, automatically configure and adjust a matching computing power resource adaptation template with the established service demand model.

[0041] According to an embodiment of the present invention, the optimal computing power resource adaptation based on the dynamic information of the target computing power resources matched by the computing power adaptation template and the isolation process are carried out simultaneously, including: Match the target computing power resources in the computing power resource pool according to the computing power adaptation template; Obtain the dynamic information of the target computing power resources, including deployment location, load status, and compatibility; Perform optimal computing power resource adaptation for the service according to the dynamic information through a preset resource scheduling model; Carry out isolation processing on the computing power resource adaptation through a preset containerization model and software-defined network model.

[0042] Among them, through a preset resource scheduling model such as a genetic algorithm model or a simulated annealing algorithm model, the dynamic deployment location, load, compatibility, and service priority of the matched target computing power resources are optimized and adapted for service resources to generate an optimal adaptation plan for resource allocation. For example, the simulated annealing algorithm performs vector encoding and fitness function establishment on the dynamic deployment location, load, compatibility, and service priority of computing power resources, and then obtains the optimal solution through iterative calculation of function differences, which is the optimal deployment plan. At the same time, it also dynamically allocates appropriate computing power resources for services through the API interface with the cloud computing platform. During the process of optimizing resource allocation, through a preset containerization model such as K8s and a software-defined network model such as SDN, isolation processing is performed on service resource allocation to create an independent operating environment, ensuring that multiple services do not interfere with each other during parallel allocation, and improving the high utilization rate and security of resources.

[0043] According to an embodiment of the present invention, the processing of the performance index data of the monitored optimal computing power resources, predicting the load trend, and obtaining the corresponding computing power resource adjustment strategy includes: Real-time monitor the operating status of the adapted optimal computing power resources and extract performance index data; The performance index data includes CPU utilization rate, memory occupancy rate, disk read and write speed, and network traffic; Process the performance index data through a preset load prediction model to obtain load trend data; Compare the load trend data with a preset load prediction threshold to obtain the corresponding computing power resource adjustment strategy.

[0044] Among them, the preset load prediction model is an operation processing model that calculates and processes the performance index data of the real-time collected operating status based on a preset algorithm. The preset algorithm can adopt algorithms such as regression algorithms, decision trees or random forest algorithms, LSTM, etc. For example, when using a linear regression algorithm, a linear relationship between CPU utilization rate, memory occupancy rate, disk read and write speed, and network traffic and the load is established, a linear equation is established to fit historical data, the weight coefficients of each index data are determined, and the future load trend is linearly predicted based on the coefficients and index values. Then, through threshold comparison and interval fitting with the preset load prediction threshold, the interval corresponding to the calculated load trend data is obtained, and the strategy corresponding to the interval is the adjustment strategy for computing power resources. For example, the threshold comparison intervals in the embodiment of this solution are [0, 0.32), [0.32, 0.48), [0.48, 0.79), [0.79, 1.0], corresponding to the first, second, third, and fourth computing power resource adjustment strategies respectively. If the threshold comparison result falls into the third interval, the computing power resources are adjusted according to the third strategy. The adjustment strategy includes increasing / decreasing the number of CPU cores, increasing / decreasing the memory capacity, or releasing idle computing power resources and returning them to the computing power resource pool.

[0045] According to an embodiment of the present invention, the processing of the corresponding computing power resource historical data in combination with the load trend data to obtain a computing power resource expansion plan, and the expansion deployment of the business computing power resources includes: Obtain the computing power resource historical data of the optimal computing power adaptation template of the business, including the average utilization rate of computing power resources and the coordination effectiveness coefficient of computing power resources; Process the average utilization rate of computing power resources and the coordination effectiveness coefficient of computing power resources in combination with the load trend data to obtain a computing power optimization degree coefficient; Query according to the computing power optimization degree coefficient through a preset computing power optimization configuration queue list to obtain a computing power resource expansion plan, including hardware upgrade or policy optimization improvement; Expand and deploy the business computing power resources according to the computing power resource expansion plan.

[0046] Among them, the average utilization rate of computing power resources is the total average utilization rate of resources of the computing power adaptation template within a preset historical time period, and the coordination effectiveness coefficient of computing power resources is an evaluation coefficient (taking values from 0 to 1) of the coordination effectiveness of the computing power resource management center platform for the computing power adaptation template within the historical time period. Then, in combination with the load trend data, a weighted calculation is performed to obtain the computing power optimization degree coefficient, and then an expansion plan is obtained through list query. Among them, the hardware upgrade includes upgrading the CPU, increasing the memory, and upgrading the storage device, and the policy optimization improvement includes optimizing the resource allocation area, adjusting the allocation priority, and improving the load balancing algorithm, so as to improve the overall performance and efficiency of computing power resources and ensure continuous satisfaction of the computing power requirements of the business.

[0047] The third aspect of the present invention provides a readable storage medium, in which a program for a method of managing and allocating computing power resources is stored. When the program for a method of managing and allocating computing power resources is executed by a processor, the steps of a method of managing and allocating computing power resources as described in any one of the above are implemented.

[0048] A method, system, and medium for managing and allocating computing power resources disclosed by the present invention integrate various computing power resources in a private network, establish a computing power resource information library, analyze computing power resource requests, establish a business demand model and optimize and match a computing power resource configuration template to obtain a computing power adaptation template, perform optimal computing power resource adaptation according to the dynamic information of the target computing power resources matched by the computing power adaptation template, and at the same time perform isolation processing, process the monitored performance index data, predict the load trend and obtain the corresponding computing power resource adjustment strategy, and optimize and expand the computing power resources according to big data analysis; thereby, by constructing a dynamic computing power resource pool, a resource scheduling model, a monitoring and adjustment, and an optimization and expansion mechanism, the efficient utilization and elastic expansion of the computing power resources of the private network are realized, and the processing efficiency and security of big data services are significantly improved.

[0049] In several embodiments provided in this application, it should be understood that the disclosed devices and methods can be implemented in other ways. The device embodiments described above are merely illustrative. For example, the division of the units is only a logical function division. In actual implementation, there may be other division methods. For example, multiple units or components can be combined, or can be integrated into another system, or some features can be ignored or not executed. In addition, the coupling, direct coupling, or communication connection between the displayed or discussed components can be through some interfaces. The indirect coupling or communication connection of devices or units can be electrical, mechanical, or other forms.

[0050] The units described above as separate components may or may not be physically separated. The components displayed as units may or may not be physical units; they can be located in one place or distributed to multiple network units; some or all of the units can be selected according to actual needs to achieve the purpose of the solution of this embodiment.

[0051] In addition, each functional unit in each embodiment of the present invention can be all integrated in a processing unit, or each unit can be separately used as a unit, or two or more units can be integrated in a unit; the above-mentioned integrated units can be implemented in the form of hardware, or in the form of a combination of hardware and software functional units.

[0052] Those of ordinary skill in the art can understand that all or part of the steps of implementing the above method embodiments can be completed by hardware related to program instructions. The foregoing program can be stored in a readable storage medium. When the program is executed, it executes the steps including the above method embodiments; and the foregoing storage medium includes: removable storage devices, read-only memory (ROM, Read-Only Memory), random access memory (RAM, Random Access Memory), magnetic disks or optical disks and other various media that can store program codes.

[0053] Alternatively, if the above-mentioned integrated units of the present invention are implemented in the form of software functional modules and sold or used as independent products, they can also be stored in a readable storage medium. Based on such an understanding, the technical solution of the embodiments of the present invention essentially or the part that contributes to the prior art can be embodied in the form of a software product. The software product is stored in a storage medium and includes several instructions for causing a computer device (which can be a personal computer, a server, or a network device, etc.) to execute all or part of the methods described in each embodiment of the present invention. And the foregoing storage medium includes: removable storage devices, ROM, RAM, magnetic disks or optical disks and other various media that can store program codes.

Claims

1. A computing power resource management and allocation method, characterized in that It includes the following steps: Integrate various computing power resources in a predetermined private network to establish a computing power resource information database; Analyze the computing power resource requests of the service, establish a service demand model, optimize and match the computing power resource configuration template, and obtain a computing power adaptation template; Perform optimal computing power resource adaptation according to the dynamic information of the target computing power resources matched by the computing power adaptation template, and perform isolation processing at the same time; Process according to the performance index data of the monitored optimal computing power resources, predict the load trend, and obtain the corresponding computing power resource adjustment strategy; Process according to the corresponding historical data of computing power resources combined with load trend data to obtain a computing power resource expansion plan, and expand and deploy the service computing power resources.

2. The computing power resource management and allocation method according to claim 1, characterized in that, The integration of various computing power resources in the predetermined private network to establish a computing power resource information database includes: Integrate various computing power resource modules in the predetermined private network to form a computing power resource pool; Perform performance evaluation and feature description on each preset computing power unit in the computing power resource pool to establish a computing power resource information database.

3. The computing power resource management and allocation method according to claim 2, characterized in that, The analysis of the computing power resource requests of the service, the establishment of a service demand model, the optimization and matching of the computing power resource configuration template, and the obtaining of a computing power adaptation template include: Analyze the computing power resource requests of the service, and extract service feature information, including service type, load characteristics, performance requirements, and service attributes; Establish a service demand model according to the service type, load characteristics, and performance requirements; Obtain multiple matching computing power resource configuration templates through the computing power resource information database according to the service attributes combined with the preset service priorities; Optimize and match the multiple computing power resource configuration templates according to the service demand model to obtain a computing power adaptation template.

4. A computing power resource management and allocation method according to claim 3, characterized in that The optimal computing power resource adaptation according to the dynamic information of the target computing power resources matched by the computing power adaptation template, and the simultaneous isolation processing include: Match the target computing power resources in the computing power resource pool according to the computing power adaptation template; Obtain the dynamic information of the target computing power resources, including deployment location, load status, and compatibility; Perform optimal computing power resource adaptation for the service according to the dynamic information through a preset resource scheduling model; Perform isolation processing on the computing power resource adaptation through a preset containerization model and software-defined network model.

5. A computing power resource management and allocation method according to claim 4, characterized in that, The processing according to the monitored performance index data of the optimal computing power resources, predicting the load trend, and obtaining the corresponding computing power resource adjustment strategy includes: Real-time monitor the operation status of the adapted optimal computing power resources, and extract performance index data; The performance index data includes CPU utilization rate, memory occupancy rate, disk read and write speed, and network traffic; Process the performance index data through a preset load prediction model to obtain load trend data; Compare the load trend data with a preset load prediction threshold to obtain the corresponding computing power resource adjustment strategy.

6. A computing power resource management and allocation method according to claim 5, characterized in that, The processing according to the corresponding historical data of computing power resources combined with load trend data to obtain a computing power resource expansion plan, and expand and deploy the service computing power resources includes: Obtain the historical data of the computing power resources of the optimal computing power adaptation template of the service, including the average utilization rate of computing power resources and the effective coefficient of computing power resource coordination; Process the average utilization rate of computing power resources and the coordination effective coefficient of computing power resources in combination with the load trend data to obtain the computing power optimization coefficient; Query through a preset computing power optimization configuration queue list according to the computing power optimization coefficient to obtain a computing power resource expansion plan, including hardware upgrade or policy optimization improvement; Expand and deploy business computing power resources according to the computing power resource expansion plan.

7. A computing power resource management and allocation system, characterized in that, Include: Computing power resource management module: Integrate various computing power resource modules in a predetermined private network to form a computing power resource pool, evaluate the performance of each preset computing power unit therein, and establish a computing power resource information database; Business demand analysis module: Analyze the computing power resource requests of the business, establish a business demand model and optimize and match the computing power resource configuration template to obtain a computing power adaptation template and obtain computing power resource allocation suggestions; Computing power resource scheduling module: Perform optimal computing power resource adaptation on the dynamic information of the target computing power resources matched by the computing power adaptation template, and perform isolation processing at the same time; Monitoring and adjustment module: Monitor the performance index data of the optimal computing power resources, predict the load trend, and obtain the corresponding computing power resource adjustment strategy; Optimization and expansion module: Perform computing power optimization degree evaluation according to the historical data of the computing power resources corresponding to the optimal computing power adaptation template in combination with the load trend data to obtain the corresponding computing power resource expansion plan, and expand and deploy the business computing power resources.

8. The computing power resource management and allocation system according to claim 7, wherein, The system further includes: a memory and a processor. The memory includes a program for a method of managing and allocating computing power resources. When the program for the method of managing and allocating computing power resources is executed by the processor, the following steps are implemented: Integrate various computing power resources in a predetermined private network and establish a computing power resource information database; Analyze the computing power resource requests of the business, establish a business demand model and optimize and match the computing power resource configuration template to obtain a computing power adaptation template; Perform optimal computing power resource adaptation on the dynamic information of the target computing power resources matched by the computing power adaptation template, and perform isolation processing at the same time; Process according to the monitored performance index data of the optimal computing power resources, predict the load trend and obtain the corresponding computing power resource adjustment strategy; Process according to the corresponding historical data of computing power resources in combination with the load trend data to obtain a computing power resource expansion plan, and expand and deploy the business computing power resources.

9. A computing power resource management and allocation system according to claim 8, characterized in that, The integrating various computing power resources in a predetermined private network and establishing a computing power resource information database includes: Integrate various computing power resource modules in a predetermined private network to form a computing power resource pool; Evaluate the performance and describe the characteristics of each preset computing power unit in the computing power resource pool, and establish a computing power resource information database.

10. A computer-readable storage medium, characterized in that, The computer-readable storage medium includes a program for a method of managing and allocating computing power resources. When the program for the method of managing and allocating computing power resources is executed by a processor, the steps of a method of managing and allocating computing power resources as described in any one of claims 1 to 6 are implemented.

Images