Intelligent scheduling management system for computing power resources

Abstract

The invention discloses a computing power resource intelligent scheduling management system, and relates to the field of computers, and the system comprises a monitoring module which is used for collecting computing power resource application information of computer equipment in real time, and monitoring computing power application risks of the computer equipment based on the computing power resource application information; the scheduling module is used for analyzing the activeness and importance of a running program on the computer equipment, selecting the running program as a computing power resource scheduling target based on the activeness and importance of the running program, and executing scheduling operation; the system can collect computing power resource application information of computer equipment in real time, monitor risks, analyze program activeness and importance, reasonably select a scheduling target for scheduling, predict a computing power shortage index, timely trigger scheduling operation according to a prediction result, timely adjust computing power resources occupied by program operation, and improve the scheduling efficiency. And the situation of insufficient computing power is avoided, so that the system is ensured to have relatively sufficient computing power resources in real time, and the operation efficiency of computer equipment is improved.

Description

Technical Field

[0001] The present invention relates to the field of computer technology, and specifically to an intelligent scheduling and management system for computing power resources. Background Art

[0002] The scheduling of computing power resources is a key operation mechanism in the intelligent era. It flexibly allocates the computing power of various computing devices according to factors such as task requirements and the status of computing power nodes. It can disassemble complex computing tasks and accurately match them to the most suitable servers, clusters, or cloud computing power, achieving efficient utilization of resources, reducing costs, and significantly improving computing efficiency.

[0003] The invention patent application with the application number 202411864033.0 discloses an intelligent computing power resource scheduling and optimization system, including a monitoring and acquisition module, a scheduling and allocation module, an optimization and prediction module, and a task management module. The monitoring and acquisition module is used to collect and monitor the operation data of computing power resources. The scheduling and allocation module dynamically schedules computing power resources based on task requirements. The optimization and prediction module predicts resource requirements based on historical data and optimizes resource allocation. The task management module is used to manage the task information of users. The monitoring and acquisition module includes a hardware monitoring unit, a resource acquisition unit, and a transmission control unit. The hardware monitoring unit is used to monitor the hardware status information of computing nodes. The resource acquisition unit is used to collect the resource usage of tasks. The transmission control unit is used to control the external transmission of monitoring and acquisition data. The scheduling and allocation module includes a resource allocation unit, a task scheduling unit, and a policy selection unit. The resource allocation unit is used to calculate the amount of resources allocated to each task. The task scheduling unit schedules tasks to corresponding nodes based on the allocated amount of resources. The policy selection unit is used to select the scheduling policy applied by the task scheduling unit. This application aims to solve the problem that "the existing technology cannot predict the change of resource requirements of tasks, resulting in the lag of resource scheduling and the need to improve the overall operation efficiency of the system".

[0004] However, relying solely on the method of predicting computing power resources obviously cannot absolutely meet the real-time and less obvious regular computing power resource requirements.

[0005] For this reason, we propose an intelligent scheduling and management system for computing power resources. Summary of the Invention

[0006] In view of the above-mentioned drawbacks of the prior art, the present invention provides an intelligent scheduling and management system for computing power resources, which can effectively solve the problems of the prior art.

[0007] To achieve the above objectives, the present invention is realized through the following technical solutions:

[0008] The present invention discloses an intelligent scheduling and management system for computing power resources, including:

[0009] A monitoring module, which is used to collect real-time information on the computing power resource application of a computer device and monitor the computing power application risk of the computer device based on the computing power resource application information; a scheduling module, which is used to analyze the activity and importance of the programs running on the computer device, select the running programs as the computing power resource scheduling targets based on the activity and importance of the running programs, and execute the scheduling operation; a prediction module, which is used to obtain in real time the computing power application risk of the computer device monitored by the monitoring module and predict the computing power shortage index of the computer device based on the computing power application risk of the computer device; a trigger module, which is used to obtain the prediction result of the computing power shortage index of the computer device in the prediction module and trigger a jump based on the prediction result to jump to the operation stage of the scheduling module; an interaction module, which is used to obtain in real time the trigger result of the operation of the trigger module, and after the trigger jump, a pop-up window for interaction is synchronously issued on the display connected to the computer device during the operation stage of the scheduling module; a feedback module, which is used to generate a system operation message and synchronously upload the operation message to a preset cloud storage space:

[0010] Further, the computing power resource application information collected by the monitoring module during operation is the programs running on the current computer device and the memory occupied by each running program respectively. A control unit and a storage unit are arranged at a lower level of the monitoring module. The control unit is used to sense in real time whether there is a newly actively running or actively ended program on the computer device. When the sensing result is yes, it triggers the monitoring module to refresh and run again, and execute the operation of monitoring the computing power application risk of the computer device. The storage unit is used to receive the computing power resource application information collected by the monitoring module during operation and store the application information;

[0011] Among them, the operation of collecting the computing power resource application information in the monitoring module is executed in real time at a frequency customized by the system end user. The initial frequency is set to 1 second / time. When the storage unit stores the computing power resource application information of the computer device, it is stored separately based on the source application of the computing power resource application information, and each piece of computing power resource application information is marked with a collection timestamp, and the computing power resource application information stored in each separately stored interval is sorted based on the marked collection timestamp.

[0012] Further, the monitoring module records the computing power application risk of the computer device monitored in the historical operation. The monitoring logic of the computing power application risk of the computer device is:

[0013]

[0014] In the formula: K is the computing power application ratio of the computer device; n is the total number of programs running on the current computer device; Q i is the memory occupied by the i-th running program; Q MAX is the maximum available memory of the computer device; σ is the correction;

[0015] Among them, Q min is the program that is not currently running and requires the least memory among the programs loaded on the computer device. For the memory required for operation, if any one or more of Formula (1) and Formula (2) hold, it indicates that there is a risk in the computing power application of the computer device. When neither Formula (1) nor Formula (2) holds, it indicates that there is no risk in the computing power application of the computer device.

[0016] Furthermore, the value of the correction σ follows:

[0017] Obtain the historical computing power resource application information of the computer device, obtain the current computing power resource application information of the computer device, and create several data sets for the programs included in each historical computing power resource application information and the current computing power resource application information. The data sets derived from the historical computing power resource application information are denoted as A1, A2, A3,..., A x , and the data set derived from the current computing power resource application information is denoted as B;

[0018] Among A1, A2, A3,..., A x Obtain the programs corresponding to the intersection, and create a set denoted as A′ based on the programs corresponding to the intersection. When the intersection A′ is an empty set, select three or more programs with the highest occurrence times in A1, A2, A3,..., A x to create a set, denoted as A′;

[0019]

[0020] In the formula: s is the reference value; q(B∩A′) is the total number of programs in the intersection; q(B∪A)′ is the total number of programs in the union;

[0021] Among them, the value of the correction σ is in the range of 0 to 1, and the reference value s is inversely proportional to the correction σ.

[0022] Furthermore, the analysis logic of the activity and importance of the programs running on the computer device in the scheduling module is expressed as:

[0023]

[0024] In the formula: F(α1) is the recommendation index when the running program α1 is selected as the computing power resource scheduling target; pα1 is the number of times the running program α1 runs in the historical computing power resource application information; p all is the number of historical computing power resource application information; t j is the continuous running time when the running program α1 runs for the jth time; t max , t min are the longest continuous running time and the shortest continuous running time of the running program α1 in the historical computing power resource application information; QMAX ′, Q MIN ′ is the maximum running occupied memory and the minimum running occupied memory of the running program α1 in the historical computing power resource application information; Q j is the average running occupied memory monitored in seconds when the running program α1 runs for the jth time; γ is the normalization factor;

[0025] Among them, represents the operation of taking the average value of , represents the operation of taking the average value of . Based on the above formula calculation, the running programs are sorted in descending order based on the recommendation index, and the running programs in the rear position in the running programs sorted in descending order are selected as the computing power resource scheduling target, which are respectively used to represent the activity and importance of the running program.

[0026] Furthermore, the prediction logic of the computer computing power shortage index in the prediction module is:

[0027]

[0028] In the formula: EL is the computer computing power shortage index; u is the historical computing power application ratio of the computer device; K v+1 , K v are the computing power application ratios of the computer device obtained for the (v + 1)th and vth times; n is the total number of programs running on the current computer; f(P i,ass ) is the decision function; P i,ass is the decision result of whether the highest correlation program of the ith program is in n;

[0029] Among them, f(P i,ass ) takes a value of 0 or 1. If the highest correlation program of the ith program is in n, then f(P i,ass ) = 0, otherwise f(P i,ass ) = 1. The larger EL is, the more shortage of computer computing power.

[0030] Furthermore, during the operation stage of the trigger module, the prediction results of the computer device computing power shortage index for the latest two times of the prediction module are cumulatively obtained, and are denoted as EL before , EL now based on the acquisition order. If EL now > EL before , then a jump is triggered, otherwise, no jump is triggered;

[0031] Among them, after each operation of the trigger module ends, the one denoted as EL before is deleted, and the one denoted as EL now is retained.

[0032] Further, the content of the interactive pop-up window issued by the operation of the interactive module is user-defined by the system-side user, and is used to prompt the user of the computer device that the computer device is about to perform a computing power resource scheduling operation based on the scheduling module;

[0033] Among them, the interactive pop-up window issued on the display connected to the computer device is cancelled from display by the user manually clicking. After the interactive pop-up window is cancelled from display, the running program that the user clicks for the first time on the computer device is excluded from the selection targets during the selection stage of the computing power resource scheduling target.

[0034] Further, the content of the system operation message includes: the monitoring result of the computing power application risk of the computer device, the prediction result of the computing power shortage index of the computer device, the number of trigger jumps triggered by the trigger module, and the cumulative number of programs selected by the scheduling module as the computing power resource scheduling target;

[0035] Among them, the system-side user or the computer device user downloads and reads the system operation message in the cloud storage space.

[0036] Further, a control unit and a storage unit are connected to the lower level of the monitoring module through wireless network interaction. The monitoring module is connected to a prediction module and a scheduling module through wireless network interaction. The prediction module is connected to a trigger module and an interactive module through wireless network interaction. The scheduling module is connected to a feedback module through wireless network interaction. The trigger module is connected to the scheduling module through wireless network interaction.

[0037] Adopting the technical solution provided by the present invention, compared with the known prior art, it has the following beneficial effects:

[0038] In the present invention, the system can collect the computing power resource application information of the computer device in real time and monitor the risks, analyze the activity and importance of the programs, reasonably select the scheduling target for scheduling. At the same time, it predicts the computing power shortage index, triggers the scheduling operation in time according to the prediction result, and will also prompt the user through a pop-up window. The system runs synchronously to generate a message to record relevant key information for the user to view. Through these functions, the system can adjust the computing power resources occupied by the program operation in time, avoid the situation of insufficient computing power, so as to ensure that the system always has relatively sufficient computing power resources and improve the operation efficiency of the computer device. BRIEF DESCRIPTION OF THE DRAWINGS

[0039] In order to more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the following will briefly introduce the drawings required for the description of the embodiments or the prior art. Obviously, the drawings in the following description are only some embodiments of the present invention. For those of ordinary skill in the art, other drawings can be obtained based on these drawings without creative efforts.

[0040] Figure 1 It is a schematic structural diagram of an intelligent scheduling and management system for computing power resources. Specific implementation manners

[0041] To make the objectives, technical solutions, and advantages of the embodiments of the present invention clearer, the technical solutions in the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings in the embodiments of the present invention. Apparently, the described embodiments are some, but not all, of the embodiments of the present invention. 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.

[0042] The following further describes the present invention with reference to embodiments.

[0043] Embodiment:

[0044] The intelligent scheduling and management system for computing power resources in this embodiment, as Figure 1 shown, includes:

[0045] A monitoring module, configured to collect in real time the application information of computing power resources of computer devices, and monitor the computing power application risks of computer devices based on the application information of computing power resources;

[0046] The application information of computing power resources collected by the monitoring module during operation is the programs running on the current computer device and the memory occupied by each running program respectively. A control unit and a storage unit are arranged at the lower level of the monitoring module. The control unit is configured to sense in real time whether there are newly actively running or actively ended programs on the computer device. When the sensing result is yes, it triggers the monitoring module to refresh and run again to perform the monitoring operation of the computing power application risks of the computer device. The storage unit is configured to receive the application information of computing power resources of the computer device collected by the monitoring module during operation and store the application information;

[0047] Among them, the operation of collecting the application information of computing power resources of computer devices in the monitoring module is executed in real time at a user-defined frequency by the system end. The initial frequency is set to 1 second / time. When the storage unit stores the application information of computing power resources of computer devices, it stores them separately based on the source applications of the application information of computing power resources. Moreover, each piece of application information of computing power resources is marked with a collection timestamp, and the application information of computing power resources stored in each separate storage interval is sorted based on the marked collection timestamp;

[0048] The monitoring module records the computing power application risks of computer devices monitored during historical operations. The monitoring logic of the computing power application risks of computer devices is:

[0049]

[0050] In the formula: K is the computing power application ratio of the computer device; n is the total number of programs running on the current computer device; Qi is the memory occupied by the i-th running program; Q MAX is the maximum available memory of the computer device; σ is the correction;

[0051] Among them, Q min is the memory required for the program that is currently not running and has the smallest memory requirement among the programs loaded on the computer device. If any one or more of Formula (1) and Formula (2) hold, it indicates that there is a risk in the computing power application of the computer device. When both Formula (1) and Formula (2) do not hold, it indicates that there is no risk in the computing power application of the computer device;

[0052] Through the above logical formula, the computing power application ratio of the computer device is calculated, so as to provide support for whether there is a risk in the computing power application of the computer device.

[0053] The value of the correction σ follows:

[0054] Obtain the historical computing power resource application information of the computer device, obtain the current computing power resource application information of the computer device, and create several data sets with the programs included in each historical computing power resource application information and the current computing power resource application information. The data sets derived from the historical computing power resource application information are denoted as A1, A2, A3,..., A x , and the data set derived from the current computing power resource application information is denoted as B;

[0055] Among A1, A2, A3,..., A x Obtain the programs corresponding to the intersection, and create a set denoted as A′ based on the programs corresponding to the intersection. When the intersection A′ is an empty set, select three or more programs with the highest occurrence times in A1, A2, A3,..., A x to create a set, denoted as A′;

[0056]

[0057] In the formula: s is the reference value; q(B∩A′) is the total number of programs in the intersection; q(B∪A)′ is the total number of programs in the union;

[0058] Among them, the value of the correction σ is in the range of 0 to 1, and the reference value s is inversely proportional to the correction σ;

[0059] Through the above logic, the value of the correction σ is restricted.

[0060] The scheduling module is used to analyze the activity and importance of the programs running on the computer device, select the running programs as the computing power resource scheduling targets based on the activity and importance of the running programs, and execute the scheduling operation;

[0061] The analysis logic of the activity and importance of the programs running on the computer device in the scheduling module is expressed as:

[0062]

[0063] In the formula: F(α1) is the recommendation index when the running program α1 is selected as the computing power resource scheduling target; pα1 is the number of times the running program α1 runs in the historical computing power resource application information; p all is the quantity of historical computing power resource application information; t j is the continuous running time when the running program α1 runs for the jth time; t max and t min are the longest and shortest continuous running times of the running program α1 in the historical computing power resource application information; Q MAX ′ and Q MIN ′ are the maximum and minimum running occupied memory of the running program α1 in the historical computing power resource application information; Q j is the average running occupied memory monitored in seconds when the running program α1 runs for the jth time; γ is the normalization factor;

[0064] Among them, represents the operation of taking the mean value of , represents the operation of taking the mean value of , and based on the above formula calculation, the running programs are sorted in descending order based on the recommendation index, and the running programs in the rear positions among the running programs sorted in descending order are selected as the computing power resource scheduling targets, are respectively used to represent the activity and importance of the running programs;

[0065] Through the above logical formula calculation, it provides a basis for selecting the running program as the computing power resource scheduling target, so as to accurately realize the selection of the running program as the computing power resource scheduling target;

[0066] The prediction module is used to obtain the computing power application risk of the computer device monitored by the monitoring module in real time, and predict the computing power shortage index of the computer device based on the computing power application risk of the computer device;

[0067] The prediction logic of the computer computing power shortage index in the prediction module is:

[0068]

[0069] In the formula: EL is the computer computing power shortage index; u is the historical computing power application ratio of the computer device; K v+1 and K v are the computing power application ratios of the computer device obtained for the (v + 1)th and vth times; n is the total number of programs running on the current computer; f(P i,ass) is the decision function; P i,ass is the determination result of whether the program with the highest relevance of the i-th program is in n;

[0070] where f(P i,ass ) takes a value of 0 or 1. If the program with the highest relevance of the i-th program is in n, then f(P i,ass ) = 0; otherwise, f(P i,ass ) = 1. The larger EL is, the more scarce the computing power of the computer device is;

[0071] Through the above logical formula calculation, it provides further operation support for the operation of the trigger module in this embodiment of the system.

[0072] The trigger module is used to obtain the prediction result of the computing power shortage index of the computer device in the prediction module, and trigger a jump based on the prediction result to jump to the operation stage of the scheduling module;

[0073] During the operation stage of the trigger module, the prediction results of the computing power shortage index of the computer device in the latest two times are cumulatively obtained, and are denoted as EL before and EL now in the order of acquisition. If EL now > EL before , then trigger a jump; otherwise, do not trigger a jump;

[0074] Among them, after each operation of the trigger module ends, the one denoted as EL before is deleted, and the one denoted as EL now is retained;

[0075] The interaction module is used to obtain the operation trigger result of the trigger module in real time, and after the trigger jump, a pop-up window for interaction is synchronously sent on the display connected to the computer device during the operation stage of the scheduling module;

[0076] The content of the pop-up window for interaction sent by the operation of the interaction module is user-defined by the system-side user, and is used to prompt the user of the computer device that the computer device is about to perform a computing power resource scheduling operation based on the scheduling module;

[0077] Among them, the pop-up window for interaction sent on the display connected to the computer device is cancelled by the user manually. After the pop-up window for interaction is cancelled, the first running program clicked by the user on the computer device is excluded from the selection targets during the selection stage of the computing power resource scheduling target;

[0078] The feedback module is used to generate a system operation message and synchronously upload the operation message to a preset cloud storage space;

[0079] The content of the system operation message includes: the monitoring results of the computing power application risk of computer devices, the prediction results of the computing power shortage index of computer devices, the number of trigger jumps triggered by the trigger module, and the cumulative number of programs selected by the scheduling module as the computing power resource scheduling target;

[0080] Among them, system - end users or computer - device users download and read the system operation message in the cloud storage space;

[0081] The lower - level of the monitoring module is connected with a control unit and a storage unit through wireless network interaction. The monitoring module is connected with a prediction module and a scheduling module through wireless network interaction. The prediction module is connected with a trigger module and an interaction module through wireless network interaction. The scheduling module is connected with a feedback module through wireless network interaction. The trigger module is interactively connected with the scheduling module through wireless network.

[0082] In this embodiment, the monitoring module runs to collect the computing power resource application information of computer devices in real - time, monitors the computing power application risk of computer devices based on the computing power resource application information. The control unit senses in real - time whether there is a newly actively running or actively ended program on the computer device. When the sensing result is yes, it triggers the monitoring module to refresh and run again to perform the computing power application risk monitoring operation of computer devices. The storage unit synchronously receives the computing power resource application information collected by the monitoring module during operation and stores the application information. The scheduling module runs later to analyze the activity and importance of the running programs on the computer device, selects the running programs as the computing power resource scheduling target based on the activity and importance of the running programs, and performs the scheduling operation. The prediction module obtains the computing power application risk of computer devices monitored by the monitoring module in real - time, predicts the computing power shortage index of computer devices based on the computing power application risk of computer devices. The trigger module further obtains the prediction result of the computing power shortage index of computer devices in the prediction module, triggers a jump based on the prediction result, and jumps to the operation stage of the scheduling module. The interaction module obtains the trigger result of the trigger module running in real - time. After the trigger jump, the scheduling module runs and synchronously issues an interactive pop - up window on the display connected to the computer device. Finally, the feedback module generates a system operation message and synchronously uploads the operation message to the preset cloud storage space.

[0083] Through the operation of the system in the above - mentioned embodiment, the stable operation of computer devices is ensured by ensuring that computer devices always have as sufficient computing power resources as possible in real - time, making the operation of computer devices more stable and the user experience better.

[0084] In summary, in the above embodiments, the system can collect the computing power resource application information of computer devices in real time, monitor risks, analyze the activity and importance of programs, reasonably select scheduling targets for scheduling. At the same time, it predicts the computing power shortage index, triggers scheduling operations in a timely manner according to the prediction results, and also prompts the user through a pop-up window. The system runs synchronously to generate messages to record relevant key information for the user to view conveniently. Through these functions, the system can timely adjust the computing power resources occupied by program operations, avoid the situation of insufficient computing power, so as to ensure that the system always has relatively sufficient computing power resources in real time and improve the operation efficiency of computer devices.

[0085] The above embodiments are only used to illustrate the technical solutions of the present invention, rather than to limit it; although the present invention has been described in detail with reference to the foregoing embodiments, those of ordinary skill in the art should understand that they can still modify the technical solutions recorded in the foregoing embodiments, or perform equivalent replacements for some of the technical features; and these modifications or replacements will not make the essence of the corresponding technical solutions deviate from the spirit and scope of the technical solutions of the embodiments of the present invention.

Claims

1. An intelligent scheduling and management system for computing power resources, characterized in that, Including: A monitoring module, which is used to collect the computing power resource application information of the computer device in real time, and monitor the computing power application risk of the computer device based on the computing power resource application information; A scheduling module, which is used to analyze the activity and importance of the programs running on the computer device, select the running programs as the computing power resource scheduling targets based on the activity and importance of the running programs, and execute the scheduling operation; A prediction module, which is used to obtain the computing power application risk of the computer device monitored by the monitoring module in real time, and predict the computing power shortage index of the computer device based on the computing power application risk of the computer device; A triggering module, which is used to obtain the prediction result of the computing power shortage index of the computer device in the prediction module, and trigger a jump based on the prediction result, and jump to the operation stage of the scheduling module; An interaction module, which is used to obtain the triggering result of the triggering module in real time. After the triggering jump, an interaction pop-up window is sent out on the display connected to the computer device synchronously during the operation stage of the scheduling module; A feedback module, which is used to generate a system operation message and upload the operation message to a preset cloud storage space synchronously.

2. The intelligent scheduling and management system for computing power resources according to claim 1, characterized in that The computing power resource application information collected by the monitoring module during operation is the programs running on the current computer device and the memory occupied by each running program. A control unit and a storage unit are arranged at the lower level of the monitoring module. The control unit is used to sense in real time whether there is a newly actively running or actively ended program on the computer device. When the sensing result is yes, it triggers the monitoring module to refresh and run again, and execute the computer device computing power application risk monitoring operation again. The storage unit is used to receive the computing power resource application information collected by the monitoring module during operation and store the application information; Among them, the operation of collecting the computing power resource application information in the monitoring module is executed in real time at a user-defined frequency by the system end user. The initial frequency is set to 1 second / time. When the storage unit stores the computing power resource application information of the computer device, it is stored separately based on the source application of the computing power resource application information. And each piece of computing power resource application information is marked with a collection timestamp, and the computing power resource application information stored in each differentiated storage interval is sorted based on the marked collection timestamp.

3. The intelligent scheduling and management system for computing power resources according to claim 2, wherein The monitoring module records the computing power application risk of the computer device monitored in historical operations. The monitoring logic of the computing power application risk of the computer device is: Where: K is the computing power application ratio of the computer device; n is the total number of programs running on the current computer device; Q i is the memory occupied by the i-th running program; Q MAX is the maximum available memory of the computer device; σ is the correction; Among them, Q min is the program that is not currently running and requires the least memory among the programs loaded on the computer device. For the memory required for operation, if any one or more of Formula (1) and Formula (2) hold, it indicates that there is a risk in the computing power application of the computer device. When neither Formula (1) nor Formula (2) holds, it indicates that there is no risk in the computing power application of the computer device.

4. The intelligent scheduling and management system for computing power resources according to claim 3, wherein The modified σ value follows: Obtain the historical computing power resource application information of the computer device, obtain the current computing power resource application information of the computer device, and create several data sets by applying the programs included in each piece of historical computing power resource application information and the current computing power resource application information. The data sets derived from the historical computing power resource application information are denoted as A1, A2, A3, ..., A x , and the data set derived from the current computing power resource application information is denoted as B; Among A1, A2, A3, ..., A x Obtain the intersection corresponding program, and create a set denoted as A' based on the intersection corresponding program. When the intersection A' is an empty set, select three or more programs with the highest occurrence times among A1, A2, A3, ..., A x to create a set, denoted as A'; In the formula: s is the reference value; q(B∩A′) is the total amount of programs in the intersection; q(B∪A)′ is the total amount of programs in the union; Among them, the modified σ value ranges from 0 to 1, and the reference value s is inversely proportional to the modified σ.

5. The intelligent scheduling and management system for computing power resources according to claim 1, wherein The analysis logic of the activity and importance of the programs running on the computer device in the scheduling module is expressed as: Where: F(α1) is the recommended index when the running program α1 selects the computing power resource scheduling target; is the number of times the running program α1 runs in the historical computing power resource application information; p all is the number of historical computing power resource application information; t j is the continuous running time when the running program α1 runs for the jth time; t max and t min are the longest and shortest continuous running times of the running program α1 in the historical computing power resource application information; Q MAX ′ and Q MIN ′ are the maximum and minimum running occupied memories of the running program α1 in the historical computing power resource application information; Q j is the average running occupied memory monitored in seconds when the running program α1 runs for the jth time; γ is the normalization factor; Among them, represents the operation of taking the average value of , represents the operation of taking the average value of . Based on the above formula calculation, the operating programs are sorted in descending order based on the recommendation index, and the post-position operating program among the operating programs sorted in descending order is selected as the computing power resource scheduling target. are respectively used to represent the activity and importance of the operating program.

6. The intelligent scheduling and management system for computing power resources according to claim 1, wherein The prediction logic of the computer computing power shortage index in the prediction module is: Where: EL is the computer computing power shortage index; u is the computing power application ratio of computer devices obtained historically; K v+1 , K v are the computing power application ratios of computer devices obtained for the (v + 1)-th and v-th times; n is the total number of programs running on the current computer; f(P i,ass ) is a decision function; P i,ass is the decision result on whether the program with the highest relevance of the i-th program is in n; where f(P i,ass ) takes a value of 0 or 1. If the program with the highest relevance of the i-th program is in n, then f(P i,ass ) = 0; otherwise, f(P i,ass ) = 1. The larger the EL, the more scarce the computing power of the computer.

7. The intelligent scheduling and management system for computing power resources according to claim 1, wherein, During the operation stage of the trigger module, the latest two prediction results of the computer device computing power shortage index obtained by the prediction module are cumulatively acquired, and are denoted as EL before and EL now in the order of acquisition. If EL now > EL before , a jump is triggered; otherwise, a jump is not triggered. Among them, after each operation of the trigger module ends, it deletes the one denoted as EL before and retains the one denoted as EL now ​ 8. The intelligent scheduling and management system for computing power resources according to claim 1, wherein The content of the interaction pop-up window sent out by the interaction module during operation is user-defined by the system end user, and is used to prompt the user of the computer device that the computer device will perform a computing power resource scheduling operation based on the scheduling module; Among them, the interactive pop-up window displayed on the monitor connected to the computer device is manually clicked by the user to cancel the display. After the interactive pop-up window is cancelled, the running program first clicked by the user on the computer device is excluded from the selection targets of the computing power resource scheduling target in the selection stage.

9. The intelligent scheduling and management system for computing power resources according to claim 1, wherein The content of the system operation message includes: the monitoring result of the computing power application risk of the computer device, the prediction result of the computing power shortage index of the computer device, the number of trigger jumps triggered by the trigger module, and the cumulative number of programs selected by the scheduling module as the computing power resource scheduling target; Among them, the system-side user or the computer device user downloads and reads the system operation message in the cloud storage space.

10. The intelligent scheduling and management system for computing power resources according to claim 1, characterized in that The lower level of the monitoring module is connected to the control unit and the storage unit through wireless network interaction. The monitoring module is connected to the prediction module and the scheduling module through wireless network interaction. The prediction module is connected to the trigger module and the interaction module through wireless network interaction. The scheduling module is connected to the feedback module through wireless network interaction. The trigger module is connected to the scheduling module through wireless network interaction.

Images