An AI-based digital collaborative cloud computing resource optimization management system

Abstract

The present invention relates to the technical field of resource management, and specifically discloses an AI-based digital collaborative cloud computing resource optimization management system. The system includes: a data recording module, which is used to record the data processing amounts of different applications in the computer during historical usage. By combining two sets of data, namely the data processing amounts of different applications in the computer during historical usage and the operation status data of different applications during current usage, it is possible to evaluate the high or low demand for processor resources of different applications at this time point. Subsequently, the processor resource allocation amounts for different applications can be dynamically adjusted in combination with the evaluation results. Since the allocation result is dynamically allocated based on the current operation status of different applications, the rationality of resource allocation can be improved, and the situation where the normal operation of the application is affected due to a sudden increase in data processing during application usage exceeding the upper limit value can be avoided, thereby improving the utilization efficiency of processor resources.

Description

Technical Field

[0001] The present invention relates to the technical field of resource management, and specifically to a digital collaborative cloud computing resource optimization management system based on AI. Background Art

[0002] Resource optimization refers to the process of improving resource utilization efficiency and benefits by reasonably allocating and effectively using resources on the premise of meeting certain conditions. Applied to actual scenarios, the most common is to optimize the processor resources of a computer to improve the utilization efficiency of the processor, reduce resource waste, and thus enhance the overall performance of the system.

[0003] Traditional computer resource optimization management systems generally release processor resources by closing unnecessary background programs to improve system performance, or adjust the resource usage limit in combination with the priority data set by users for different applications and the development data of the applications to prevent a single application from occupying too many processor resources, thereby achieving the resource optimization management of the processor.

[0004] In the prior art, traditional computer resource optimization management systems generally adjust the upper limit value of resource usage according to the priority data set by users for different applications and the development data of the applications. However, since the data processing volume of an application will increase significantly during use, once it exceeds the upper limit value, it will affect the normal operation of the application, resulting in uneven distribution of processor resources and reducing the utilization efficiency of processor resources. Summary of the Invention

[0005] The purpose of the present invention is to provide a digital collaborative cloud computing resource optimization management system based on AI, and solve the following technical problems:

[0006] How to dynamically adjust the processor resource allocation amount of different applications to improve the utilization efficiency of processor resources.

[0007] The purpose of the present invention can be achieved through the following technical solutions:

[0008] A digital collaborative cloud computing resource optimization management system based on AI, the system includes:

[0009] A data recording module, used to record the data processing volume of different applications in the computer during historical use;

[0010] A data collection module, used to collect the operation status data of different applications when the computer is in current use;

[0011] A cloud computing module, used to calculate the processor resource demand index of different applications at different time points during the current use process by combining the data collected in the data collection module and the data in the data recording module;

[0012] An AI evaluation module, which is used to evaluate the high and low of the processor resource requirements of different applications at this time point in combination with the calculation results in the cloud computing module;

[0013] A data decision-making module, which is used to make decisions on the resource allocation of different applications in combination with the evaluation results of the AI evaluation module;

[0014] A resource allocation module, which is used to allocate the processor resource amounts of different applications in combination with the decision results of the data decision-making module and the calculation results in the cloud computing module.

[0015] Furthermore, the data collected by the data collection module includes:

[0016] The memory occupancy, computing intensity, usage duration, and click times of different applications in the computer.

[0017] Furthermore, the calculation process in the cloud computing module includes:

[0018] Through the formula Calculate to obtain the coefficient of variation of the data processing volume of the th application in the past period of time ;

[0019] Among them, n is the number of times the application is used in the past period of time, and ,, is the total data processing volume of the th application for the i-th time in the past period of time, is the total number of minutes when the th application is used for the i-th time in the past period of time, is the average value of all , is the maximum value among all , is the minimum value among all , is a proportionality coefficient, which is set by empirical fitting.

[0020] Furthermore, the calculation process in the cloud computing module also includes:

[0021] By assigning a value to the coefficient of variation of the data processing volume of the th application in the past period of time to generate a weight influence value of the th application that is between 1 and 5 and decreases as the coefficient of variation increases;

[0022] Among them, the coefficient of variation of the data processing volume of the th application in the past period of time The weight influence value corresponding to the th application is set to .

[0023] Furthermore, the calculation process in the cloud computing module further includes:

[0024] Calculating the processor resource demand index of the th application at the y-th time point during the current usage process through the formula ;

[0025] Among them, is the total number of data collections at the y-th time point during the current usage process, is the data processing volume of the th application at the y-th time point during the current usage process, is 's standard value, is the memory occupancy of the th application at the y-th time point during the current usage process, is 's standard value, is the computing intensity influence coefficient of the th application, set by empirical fitting, is the usage duration of the th application at the y-th time point during the current usage process, is the total usage duration of the computer at the y-th time point, is the number of clicks of the th application at the y-th time point during the current usage process, is 's standard value.

[0026] Furthermore, the evaluation process of the AI evaluation module includes:

[0027] By comparing the processor resource demand index of the th application at the y-th time point during the current usage process with the preset processor resource demand index threshold range ;

[0028] If , the system determines that the application requires less processor resources;

[0029] If , the system determines that the application requires medium processor resources;

[0030] If , the system determines that the application requires a lot of processor resources. ​

[0031] Furthermore, the decision-making process of the data decision-making module includes:

[0032] When it is determined that the application requires less processor resources, it is defined as a low-demand application, and based on the amount of processor resources currently occupied by the application, the amount of processor resource allocation is reduced;

[0033] When it is determined that the application requires medium processor resources, the amount of processor resource allocation is not adjusted;

[0034] When it is determined that the application requires a large amount of processor resources, it is defined as a high-demand application, and based on the amount of processor resources currently occupied by the application, the amount of processor resource allocation is increased.

[0035] Furthermore, the adjustment process of the resource allocation module includes:

[0036] When it is determined that the th application requires less processor resources;

[0037] The low-demand adjusted processor resource allocation amount of the th application at the yth time point is obtained by calculating through the formula ; ;

[0038] When it is determined that the th application requires a large amount of processor resources;

[0039] The high-demand adjusted processor resource allocation amount of the th application at the yth time point is obtained by calculating through the formula ; ;

[0040] Among them, is the low-demand preset processor resource allocation amount of the th application, is the high-demand preset processor resource allocation amount of the th application, is the adjustment coefficient comparison table function, which is fitted and set according to the influence degree of the value of in the empirical data on the processor resource allocation amount.

[0041] Advantages of the present invention:

[0042] (1) By combining two sets of data, namely the data processing volume of different applications in the computer during historical usage and the running state data of different applications during current usage, the present invention can evaluate the high or low demand for processor resources of different applications at this time point. Subsequently, the allocation volume of processor resources for different applications can be dynamically adjusted in combination with the evaluation results. Since the allocation result is dynamically allocated based on the current running state of different applications, the rationality of resource allocation can be improved, and the situation where the normal operation of an application is affected due to a sudden increase in data processing during application usage exceeding the upper limit value can be avoided, thereby improving the utilization efficiency of processor resources.

[0043] (2) By comparing the processor resource demand index of the th application at the y - th time point during current usage with the preset processor resource demand index threshold range Through this comparison method, it is possible to determine the amount of processor resources required by different applications, thereby providing accurate data for subsequent decisions on how to adjust the resource allocation of the application, ensuring the accuracy of the decision result, and thus realizing the optimal management of processor resources and improving the utilization rate of the processor.

[0044] (3) By combining the comparison result of the processor resource demand index of the th application at the y - th time point during current usage with the preset processor resource demand index threshold range accurate decisions can be made on the resource allocation methods for different applications, including reducing the processor resource allocation volume for low - demand applications and increasing the processor resource allocation volume for high - demand applications. By dynamically adjusting the processor resource allocation volume for different applications, it is possible to ensure the running stability of high - demand applications while the low - demand applications are running normally, thereby realizing the optimal allocation of processor resources. BRIEF DESCRIPTION OF THE DRAWINGS

[0045] The present invention will be further described below in conjunction with the accompanying drawings.

[0046] Figure 1 is a schematic block diagram of a digital collaborative cloud computing resource optimization management system based on AI in the present invention. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0047] The technical solutions in the embodiments of the present invention will be clearly and completely described below in conjunction with the accompanying drawings in the embodiments of the present invention. Obviously, the described embodiments are only a part of the embodiments of the present invention, rather than all of the embodiments. All other embodiments obtained by those of ordinary skill in the art based on the embodiments of the present invention without creative efforts shall fall within the protection scope of the present invention.

[0048] Please refer to Figure 1 As shown, in one embodiment, the present application provides an AI-based digital collaborative cloud computing resource optimization management system, and the system includes:

[0049] A data recording module, configured to record the data processing volume of different applications in the computer during historical usage;

[0050] A data collection module, configured to collect the operation status data of different applications when the computer is in current use;

[0051] A cloud computing module, configured to calculate the processor resource demand index of different applications at different time points during the current usage by combining the data collected in the data collection module with the data in the data recording module;

[0052] An AI evaluation module, configured to evaluate the high or low of the processor resource demand of different applications at this time point by combining the calculation results in the cloud computing module;

[0053] A data decision module, configured to make decisions on the resource allocation of different applications by combining the evaluation results of the AI evaluation module;

[0054] A resource allocation module, configured to allocate the processor resource amount of different applications by combining the decision results of the data decision module and the calculation results in the cloud computing module;

[0055] Through the above technical solution, this example provides a data recording module, configured to record the data processing volume of different applications in the computer during historical usage. When the computer is running, first, the data collection module collects the operation status data of different applications when the computer is in current use. Then, the cloud computing module can calculate the processor resource demand index of different applications at different time points during the current usage by combining the data collected in the data collection module with the data in the data recording module. And the AI evaluation module evaluates the high or low of the processor resource demand of different applications at this time point by combining the calculation results in the cloud computing module. After obtaining the evaluation results, the data decision module can make decisions on the resource allocation of different applications by combining the evaluation results of the AI evaluation module. And the resource allocation module allocates the processor resource amount of different applications by combining the decision results of the data decision module and the calculation results in the cloud computing module;

[0056] By setting it this way, in this example, by combining the data processing volume of different applications in the computer during historical use and the running state data of different applications during current use, the high and low demand for processor resources of different applications at this time point can be evaluated. Then, the processor resource allocation amount for different applications can be dynamically adjusted in combination with the evaluation results. Since the allocation result is dynamically adjusted based on the current running state of different applications, the rationality of resource allocation can be improved, and the situation where the normal operation of an application is affected due to a sudden increase in data processing during the use of the application exceeding the upper limit value can be avoided, thereby improving the utilization efficiency of processor resources.

[0057] The data collected by the data collection module includes:

[0058] The memory occupancy, computational intensity, usage duration, and click count of different applications in the computer;

[0059] Through the above technical solution, this example provides the data collected by the data collection module, including the memory occupancy, computational intensity, usage duration, and click count of different applications in the computer. Applications with high memory occupancy require more processor resources to ensure the stability of operation, and high-computational-intensity applications also require more processor resources to process a large amount of data and complex calculations. Therefore, this data can directly reflect the demand for processor resources of different applications, and the usage duration and click count of different applications can reflect whether different applications are background applications. Background applications can have their processor resource amount reduced, and the foreground applications can be preferentially supplied, thereby realizing the optimal management of processor resources.

[0060] The calculation process in the cloud computing module includes:

[0061] Through the formula Calculate to obtain the coefficient of variation of the data processing volume of the th application in the past period of time ;

[0062] Among them, n is the number of times the application has been used in the past period of time, and ,, is the total data processing volume of the th application for the i-th time in the past period of time, is the total number of minutes of the th application for the i-th time in the past period of time, is the average value of all , is the maximum value among all , is the minimum value among all , is the proportionality coefficient, which is set by empirical fitting;

[0063] Through the above technical solution, this example provides the coefficient of variation of the data processing volume of the th application in the past period of time, which can be obtained by calculation through the formula . This data can reflect the fluctuation of the data processing volume of the th application in the past period of time. Since the background applications in the computer are generally in the "sleep" state and their data processing volume will only increase when updated or optimized, the fluctuation of their data processing volume in the past period of time will be relatively large. While the data processing volume of the applications being used by the user is relatively stable during use, so the fluctuation of their data processing volume in the past period of time will be relatively small. By setting like this, it is possible to determine whether the application is a background application or an application being used by the user according to the coefficient of variation of the data processing volume of the th application in the past period of time, thus providing data support for subsequent adjustment of the resource allocation of the processor to ensure the rationality of resource allocation.

[0064] The calculation process in the cloud computing module further includes:

[0065] By assigning a value to the coefficient of variation of the data processing volume of the th application in the past period of time, a weight influence value of the th application located between 1 and 5 and decreasing as the coefficient of variation increases is generated;

[0066] Among them, the weight influence value of the th application corresponding to the coefficient of variation of the data processing volume of the th application is set to ; ;

[0067] Through the above technical solution, this example provides a process of assigning a value to the coefficient of variation of the data processing volume of the th application in the past period of time;

[0068] As an embodiment, the value standard of

[0069]

[0070] It should be noted that the larger the coefficient of variation of the data processing volume of the th application in the past period of time, the smaller the corresponding weight influence value of the th application, because The greater the fluctuation value of the data processing volume of an application over a period of time in the past, it indicates that the application is generally in a "dormant" state, and its data processing volume will only increase during updates or optimizations. The resource allocation of its processor can be reduced. On the contrary, the coefficient of variation of the data processing volume of the th application over a period of time in the past is smaller, the greater the weight influence value of the corresponding th application. This is because the smaller the fluctuation value of the data processing volume of the

[0071] th application over a period of time in the past, it means that the data processing volume of the application during use is relatively stable, and it is an application being used by the user. The resource allocation of its processor needs to be increased to ensure the stability of the application.

[0072] The calculation process in the cloud computing module further includes: Calculating to obtain the processor resource demand index of the th application at the y-th time point during the current usage process ;

[0073] wherein, is the total number of data acquisitions at the y-th time point during the current usage process, is the data processing volume of the th application at the y-th time point during the current usage process, is 's standard value. The above standard value can be selected and set according to the allowable error in the empirical data, is the memory occupancy of the th application at the y-th time point during the current usage process, is 's standard value. The above standard value can be selected and set according to the allowable error in the empirical data, is the computational intensity influence coefficient of the th application, which is set by empirical fitting, is the usage duration of the th application at the y-th time point during the current usage process, is the total usage duration of the computer at the y-th time point, is the number of clicks of the th application at the y-th time point during the current usage process, is 's standard value. The above standard value can be selected and set according to the allowable error in the empirical data;

[0074] Through the above technical solution, this example provides the The processor resource requirement index of an application at the y-th time point , can be obtained by the formula . Obviously, the larger the data processing volume and memory occupancy of the -th application at the y-th time point during the current usage, the shorter the usage duration of the -th application at the y-th time point during the current usage, and the more times the -th application has been clicked from the start of use to the y-th time point during the current usage, then the processor resource requirement index of the -th application at the y-th time point during the current usage is larger, indicating that the application is a currently used application and requires more processor resource allocation to improve and ensure the running stability of the application. On the contrary, the smaller the data processing volume and memory occupancy of the -th application at the y-th time point during the current usage, the longer the usage duration of the -th application at the y-th time point during the current usage, and the fewer times the -th application has been clicked from the start of use to the y-th time point during the current usage, then the processor resource requirement index of the -th application at the y-th time point during the current usage is smaller, indicating that the application is a background-running application and the processor resource allocation can be appropriately reduced;

[0075] Through this calculation method, the required resource amounts of different applications can be analyzed, so as to provide accurate data support for subsequent resource allocation to different applications to ensure the rationality of resource allocation.

[0076] The evaluation process of the AI evaluation module includes:

[0077] By comparing the processor resource requirement index of the -th application at the y-th time point during the current usage with the preset processor resource requirement index threshold range ;

[0078] If , the system determines that the application requires less processor resources;

[0079] If , the system determines that the application requires medium processor resources;

[0080] If , the system determines that the application requires a lot of processor resources;

[0081] Through the above technical solution, in this example, by comparing the The processor resource requirement index of an application at the y-th time point is compared with a preset threshold range of the processor resource requirement index Through this comparison method, it is possible to judge the amount of processor resources required by different applications, so as to provide accurate data for subsequent decisions on how to adjust the resource allocation of the application, ensure the accuracy of the decision result, and thus realize the optimal management of the processor resources and improve the utilization rate of the processor.

[0082] The decision-making process of the data decision-making module includes:

[0083] When it is judged that the processor resources required by the application are small, it is defined as a low-demand application, and on the basis of the current amount of processor resources occupied by the application, the amount of processor resource allocation is reduced;

[0084] When it is judged that the processor resources required by the application are medium, the amount of processor resource allocation is not adjusted;

[0085] When it is judged that the processor resources required by the application are large, it is defined as a high-demand application, and on the basis of the current amount of processor resources occupied by the application, the amount of processor resource allocation is increased;

[0086] Through the above technical solution, in this example, by combining the processor resource requirement index of the application at the y-th time point during the current usage process and the preset threshold range of the processor resource requirement index According to the comparison result, accurate decisions can be made on the resource allocation methods of different applications, including reducing the processor resource allocation of low-demand applications and increasing the processor resource allocation of high-demand applications. By dynamically adjusting the processor resource allocation of different applications, it is possible to ensure the operation stability of high-demand applications while the low-demand applications are running normally, thus realizing the optimal allocation of processor resources.

[0087] The adjustment process of the resource allocation module includes:

[0088] When it is judged that the processor resources required by the application are small;

[0089] The low-demand adjusted processor resource allocation amount of the application at the y-th time point is calculated through the formula ; for the application

[0090] When it is judged that the processor resources required by the application are large;

[0091] Through the formula Calculate and obtain the processor resource allocation volume after high-demand adjustment for the th application at the y-th time point ;

[0092] Wherein, is the preset processor resource allocation volume for low demand of the th application, is the preset processor resource allocation volume for high demand of the th application, is the adjustment coefficient comparison table function, which is fitted and set according to the influence degree of the value in the empirical data on the processor resource allocation volume;

[0093] Through the above technical solution, this example provides the processor resource allocation volume after adjustment for the th low-demand application at the y-th time point and the processor resource allocation volume after adjustment for the th high-demand application at the y-th time point , which can be calculated respectively through formula and formula . Through this calculation method, in combination with the processor resource demand index of the th application at the y-th time point during the current use process , since this data is calculated based on diversified data, the accuracy of this data is relatively high. Based on this data, the processor resources of different applications are allocated, which can improve the accuracy of the allocation result, thereby realizing the optimized management of processor resources.

[0094] The above has described an embodiment of the present invention in detail, but the content described is only a preferred embodiment of the present invention and cannot be considered as limiting the scope of implementation of the present invention. All equivalent changes and improvements made according to the scope of the present invention application shall still fall within the scope covered by the patent of the present invention.

Claims

1. A digital collaborative cloud computing resource optimization management system based on AI, characterized in that: The system comprises: A data recording module is used to record the data processing volume of different applications in the computer during historical use; A data collection module is used to collect the running status data of different applications when the computer is currently in use; The cloud computing module is used to combine the data collected in the data collection module with the data in the data recording module to calculate the processor resource demand index of different applications at different time points during the current use process; The AI ​​evaluation module is used to evaluate the processor resource requirements of different applications at that time point based on the calculation results in the cloud computing module; The data decision module is used to make decisions on resource allocation for different applications based on the evaluation results of the AI ​​evaluation module; The resource allocation module is used to allocate processor resources to different applications based on the decision results of the data decision module and the calculation results of the cloud computing module; The computing process in the cloud computing module includes: By formula Calculate the first The dispersion coefficient of the data processing volume of an application in the past period of time ; Where n is the number of times the application has been used in the past period of time, and , For the past period of time The total data processing volume used by the application for the i-th time, For the past period of time The total number of minutes when the app is used for the i-th time, For all The average value of For all The maximum value in For all The minimum value in is the proportionality coefficient, which is set based on empirical fitting; The computing process in the cloud computing module also includes: Through the The dispersion coefficient of the data processing volume of an application in the past period of time Assign values ​​to generate the first value between 1 and 5, which decreases as the discrete coefficient increases. The weighted impact value of each application; Among them, the The dispersion coefficient of the data processing volume of an application in the past period of time The corresponding The weighted impact value of each application is set as ; The computing process in the cloud computing module also includes: By formula Calculate the current usage process The processor resource demand index of an application at time point y ; in, is the total number of data collection times at the yth time point during the current use process, For the current use process The amount of data processed by an application at the yth time point, for The standard value of For the current use process The memory usage of an application at the yth time point, for The standard value of For the The computational intensity of each application is set based on empirical fitting. For the current use process The usage time of an application at the yth time point, is the total usage time of the computer at the yth time point, For the current use process The number of times an application is clicked at the yth time point, for The standard value of The evaluation process of the AI ​​evaluation module includes: By using the current process The processor resource demand index of an application at time point y The preset processor resource demand index threshold range Make a comparison; like , the system determines that the application requires less processor resources; like , the system determines that the processor resources required by the application are medium; like ,The system determines that the application requires more processor resources; The decision-making process of the data decision module includes: When it is determined that the application requires less processor resources, it is defined as a low-demand application, and based on the amount of processor resources currently occupied by the application, the amount of processor resources allocated is reduced; When it is determined that the processor resources required by the application are medium, the processor resource allocation amount is not adjusted; When it is determined that the application requires a lot of processor resources, it is defined as a high-demand application, and based on the amount of processor resources currently occupied by the application, the amount of processor resources allocated is increased; The adjustment process of the resource allocation module includes: When judging When an application requires less processor resources; By formula Calculate the yth time point The amount of processor resources allocated after the application has low demand adjustment ; When judging When an application requires more processor resources; By formula Calculate the yth time point Processor resource allocation after high demand adjustment for each application ; in, For the The minimum required preset processor resource allocation for each application, For the Preset processor resource allocation for high demand applications. To adjust the coefficient comparison table function, according to the empirical data The influence of the numerical value on the processor resource allocation is set appropriately.

2. According to claim 1, the AI-based digital collaborative cloud computing resource optimization management system is characterized in that: The data collected in the data collection module includes: The memory usage, computing intensity, usage time and number of clicks of different applications on the computer.

Images