Data interaction scheduling platform based on cloud computing

By preprocessing historical and real-time data, establishing a data demand prediction model and calculating device load, the problem of low interactive scheduling accuracy caused by simple priority division in the existing technology is solved, and real-time prediction and priority of data transmission requirements are realized The accuracy of level division improves the accuracy of data interaction scheduling and resource utilization efficiency.

CN118963956AInactive Publication Date: 2024-11-15CHINA YANGTZE POWER
View PDF 0 Cites 2 Cited by

Patent Information

Application Number
CN202411084749.9
Authority / Receiving Office
CN · China
Patent Type
Applications(China)
Current Assignee / Owner
Filing Date
2024-08-08
Publication Date
2024-11-15
Estimated Expiration
Not applicable · inactive patent

AI Technical Summary

Technical Problem

The existing data interactive scheduling platform is relatively simple in prioritization, resulting in low accuracy of interactive scheduling and inability to predict data transmission requirements in real time, resulting in inconvenient priority division.

Method used

By collecting historical data and real-time data for preprocessing, a data demand prediction model is established, the equipment load is calculated, the equipment is load-based, the prediction results and equipment performance are prioritized, and data interactive scheduling is performed through the priority calculation model.

Benefits of technology

It realizes the accuracy of real-time prediction and priority division of data transmission requirements, improves the accuracy of data interaction scheduling, reduces the occurrence of scheduling errors, and brings better resource utilization and commercial value.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN118963956A_ABST
    Figure CN118963956A_ABST
Patent Text Reader

Abstract

The invention discloses a data interaction scheduling platform based on cloud computing, and relates to the technical field of data scheduling, and the platform comprises the steps: collecting historical data and real-time data, and carrying out the preprocessing of the data; obtaining a data task based on the preprocessed data, calculating a transmission demand of the data based on a task index, and predicting the transmission demand of the data; based on the data prediction result, calculating the current equipment load capacity, and performing priority division on the data; and performing interactive scheduling on the data based on priority division. According to the method, the data demand prediction model is established, so that the transmission demand of the data can be predicted in real time, the load indexes of the equipment are calculated based on the data transmission demand prediction quantity and the initialization performance of the equipment, and the initial priority of the data is set according to the load indexes of all the equipment; therefore, the priority division is more convenient and accurate, the interactive scheduling accuracy of the data is higher, scheduling errors are not easy to generate, and a better use prospect is brought.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] The present invention relates to the technical field of data scheduling, and in particular to a data interaction scheduling platform based on cloud computing. Background Art

[0002] The data interaction scheduling platform is a system used to manage and optimize data transmission and processing in a cloud environment. It combines cloud computing, data management and processing, scheduling optimization, and network technology. Through virtualization, physical computing resources are abstracted into virtual resources, allowing users to obtain and use computing resources on demand. It can automatically adjust the allocation and release of resources according to actual needs to cope with changes in data interaction load. According to the characteristics of the task and the status of cloud resources, it automatically schedules and manages the task execution order and resource allocation to optimize the overall system performance and resource utilization.

[0003] During use, the existing data interaction scheduling platform has a relatively simple division of priorities, resulting in a relatively low accuracy rate of data interaction scheduling, prone to scheduling errors, and unable to predict data transmission requirements in real time, resulting in inconvenience in priority division. Summary of the invention

[0004] In order to solve the above technical problems, a data interaction scheduling platform based on cloud computing is provided. This technical solution solves the above problems.

[0005] In order to achieve the above purpose, the technical solution adopted by the present invention is:

[0006] A data interactive scheduling platform based on cloud computing, comprising:

[0007] S1: Collect historical data and real-time data and pre-process the data;

[0008] S2: Based on the preprocessed data, obtain the data task, calculate the data transmission demand based on the task indicators, and predict the data transmission demand;

[0009] S3: Based on the data prediction results, calculate the current device load and prioritize the data;

[0010] S4: Based on the priority division, data can be interactively scheduled.

[0011] Preferably, historical data and real-time data are collected, and preprocessing the data specifically includes:

[0012] Get data from the API interface and send the data to the cloud platform;

[0013] Identify and process missing values ​​in the data through filling methods, detect and remove duplicate data records in the data set, and normalize the data;

[0014] The formula for data normalization is:

[0015]

[0016] in, is the normalized value, g is the value of the original data, ming is the minimum value in the data set g, and maxg is the maximum value in the data set g.

[0017] Preferably, S2 specifically includes:

[0018] Based on the preprocessed historical operation data, classify and label the preprocessed data and identify the data tasks;

[0019] Predict data needs based on data task indicators;

[0020] Assume that there are several data tasks that need to be analyzed and supported;

[0021] For each task, key factors are defined, including: data volume requirements, data update frequency, data quality requirements and data packet loss rate;

[0022] Build a data demand forecasting model based on the key factors of each task;

[0023] The data demand prediction model formula is:

[0024] GF=(β0+DQ1×β1+DQ2×β2+DQ i ×β n )

[0025] Among them, DQ i is the i-th key factor parameter, n is the total number of key factors, β0, β1, β2, β n are regression coefficients, and GF represents the predicted data transmission demand.

[0026] Preferably, the preprocessed data is obtained, classified and marked, and the task of identifying the data specifically includes:

[0027] Based on the preprocessed historical operation data, task preference marking is performed to obtain the operation data preference classification.

[0028] Get the running task set;

[0029] Construct the original data set based on the running task set and divide it into training set and validation set;

[0030] Based on the preferred feature data and task label data in the training set, the corresponding decision tree model is trained to perform task classification;

[0031] The decision tree model formula is:

[0032] Where D is the Gini index of the data set, K is the total number of categories, PK is the proportion of samples belonging to category K in set D, is the predicted category of x;

[0033] Select the feature with the smallest Gini index as the split node;

[0034] Use the accuracy evaluation formula to evaluate the model performance, and adjust the feature selection of the model based on the accuracy results.

[0035] Preferably, the accuracy evaluation formula is used to evaluate the model performance, and according to the accuracy result, the feature selection of the model is adjusted, specifically including:

[0036] Based on the calculation results of the decision tree model, calculate the accuracy of the model;

[0037] The accuracy evaluation formula is:

[0038]

[0039] Among them, A is the accuracy, CP is the number of correct predictions, and N is the total number of samples;

[0040] Based on the accuracy calculation results, adjust the feature selection of the model;

[0041] When the accuracy is lower than the preset value, the system will automatically adjust the feature selection of the model and automatically select features with a small Gini index.

[0042] Preferably, S3 specifically includes:

[0043] Initialize the performance of the device and obtain the initialization performance indicators of the device;

[0044] Obtaining data transmission demand forecast;

[0045] Calculate the load index of the device based on the predicted data transmission demand and the initialization performance of the device;

[0046] The equipment load index calculation formula is: F 负 =GF×F J

[0047] Among them, F 负 is the load index of the equipment, GF is the predicted data transmission demand, and F J is the initialization performance of the Jth device;

[0048] Design a priority calculation model based on all equipment load indicators;

[0049] Based on the priority calculation model, the initial priority of the data is calculated;

[0050] Based on the priority calculation results, the interactive scheduling of data is allocated;

[0051] The priority calculation formula is:

[0052]

[0053] Among them, Y jt is the priority of the jth device at time t, p is the total number of hysteresis learning stages, m is the total number of devices, and α j is the j-th order autoregressive coefficient, F Jt-i is the load forecast of the jth device at time ti;

[0054] ∈ t is the error term of the t-th unit time.

[0055] Preferably, S4 specifically includes:

[0056] Categorize and label data according to identified priority criteria;

[0057] Priority scheduling based on task tags;

[0058] After the scheduling is implemented, the execution process of the task is monitored and feedback is provided;

[0059] Based on the monitoring and feedback results, adjustments and optimizations are made, including re-evaluation of priority criteria.

[0060] Preferably, based on the monitoring and feedback results, the adjustment and optimization specifically include:

[0061] After the scheduling is implemented, during the task execution process, the scheduling progress and resource usage of the task are continuously monitored, and task execution efficiency data feedback is collected. The task execution efficiency includes: execution time, resource utilization and satisfaction;

[0062] Analyze the efficiency of task execution and resource utilization based on the collected monitoring data;

[0063] If monitoring data shows that the actual priority of data tasks does not match expectations, re-evaluate and adjust the priority criteria.

[0064] Preferably, monitoring data shows that the actual priority of the data task does not match the expected one, and re-evaluation and adjustment of the priority mark are carried out.

[0065] The specific requirements include:

[0066] Acquire monitoring data of task execution, the monitoring data including task execution time, resource consumption, and order of task completion;

[0067] Determine whether the currently used priority criteria are assigned in order;

[0068] Adjust resource allocation strategies based on the re-evaluated priority criteria based on monitoring results.

[0069] Preferably, adjusting the resource allocation strategy according to the monitoring results and the re-evaluated priority criteria specifically includes: manually updating the priority field in the scheduling system to complete the reallocation of the priority division strategy;

[0070] The scheduling and monitoring process is restarted based on the adjusted priority criteria.

[0071] Implement updated prioritization criteria and continually monitor performance of tasks.

[0072] Compared with the prior art, the beneficial effects of the present invention are: it can establish a data demand prediction model based on the task indicators of the data, predict the data demand, and thus predict the data transmission demand in real time, calculate the load indicator of the equipment based on the predicted amount of data transmission demand and the initialization performance of the equipment, and set the initial priority of the data based on all the equipment load indicators, thereby making the division of priorities more convenient and accurate, resulting in a higher accuracy rate in the interactive scheduling of data, less prone to scheduling errors, and bringing better usage prospects and commercial value. BRIEF DESCRIPTION OF THE DRAWINGS

[0073] Figure 1 is a flow chart of the steps of the present invention;

[0074] Figure 2 It is a flow chart of S1 in the present invention;

[0075] Figure 3 is a flow chart of S2 in the present invention;

[0076] Figure 4 is a flow chart of S3 in the present invention;

[0077] Figure 5 This is a flow chart of S4 in the present invention. DETAILED DESCRIPTION

[0078] The following description is used to disclose the present invention so that those skilled in the art can implement the present invention. The preferred embodiments described below are only examples, and those skilled in the art may think of other obvious variations.

[0079] Reference Figure 1 As shown, a data interaction scheduling platform based on cloud computing includes:

[0080] S1: Collect historical data and real-time data and pre-process the data;

[0081] S2: Based on the preprocessed data, obtain the data task, calculate the data transmission demand based on the task indicators, and predict the data transmission demand;

[0082] S3: Based on the data prediction results, calculate the current device load and prioritize the data;

[0083] S4: Based on the priority division, data can be interactively scheduled.

[0084] Reference Figure 2 As shown, historical data and real-time data are collected, and data preprocessing is performed including:

[0085] Get data from the API interface and send the data to the cloud platform;

[0086] Identify and process missing values ​​in the data through filling methods, detect and remove duplicate data records in the data set, and normalize the data;

[0087] The formula for data normalization is:

[0088]

[0089] in, is the normalized value, g is the value of the original data, ming is the minimum value in the data set g, and maxg is the maximum value in the data set g;

[0090] It should be further explained that: for numerical data, the mean, median and mode of the overall data are used for filling;

[0091] For time series data, fill in with the previous or next non-missing value;

[0092] For continuous numeric data, use interpolation methods such as linear interpolation to fill in.

[0093] During the filling process, record the filling strategy and method.

[0094] Reference Figure 3 As shown, S2 specifically includes:

[0095] Obtain preprocessed data, classify and label the preprocessed data, and identify the data tasks;

[0096] Predict data needs based on data task indicators;

[0097] Assume that there are several data tasks that need to be analyzed and supported;

[0098] For each task, key factors are defined, including: data volume requirements, data update frequency, data quality requirements and data packet loss rate;

[0099] Based on the key factors of each task, determine the specific goals and business requirements of the task, understand the core driving factors of data demand, and establish a data demand prediction model;

[0100] The data demand prediction model formula is:

[0101] GF=(β0+DQ1×β1+DQ2×β2+DQ i ×β n )

[0102] Among them, DQ i is the i-th key factor parameter, n is the total number of key factors, β0, β1, β2, β n are regression coefficients, and GF represents the predicted data transmission demand.

[0103] It should be specifically stated that the key factors include but are not limited to: data volume requirements, data update frequency, data quality requirements, data packet loss rate, data task volume, data processing methods, data update methods, technical requirements and evaluation mechanisms;

[0104] Obtain preprocessed data, classify and label the preprocessed data, and identify the data tasks specifically including:

[0105] Based on the preprocessed historical operation data, task preference marking is performed to obtain the operation data preference classification.

[0106] Get the running task set;

[0107] Construct the original data set based on the running task set and divide it into training set and validation set;

[0108] Based on the preferred feature data and task label data in the training set, the corresponding decision tree model is trained to perform task classification;

[0109] The decision tree model formula is:

[0110] Where D is the Gini index of the data set, K is the total number of categories, PK is the proportion of samples belonging to category K in set D, is the predicted category of x;

[0111] Select the feature with the smallest Gini index as the split node;

[0112] Use the accuracy evaluation formula to evaluate the model performance, and adjust the feature selection of the model based on the accuracy results.

[0113] What needs to be explained specifically is:

[0114] First, randomize the samples in the data set to ensure that the sample distribution of the training set and the validation set is similar. Divide the constructed original data set into a training set and a validation set, using 70% of the data as the training set and 30% of the data as the validation set.

[0115] Among them, the accuracy evaluation formula is used to evaluate the model performance. According to the accuracy result, the feature selection of the model is adjusted, including:

[0116] Based on the calculation results of the decision tree model, calculate the accuracy of the model;

[0117] The accuracy evaluation formula is:

[0118]

[0119] Among them, A is the accuracy, CP is the number of correct predictions, and N is the total number of samples;

[0120] Based on the accuracy calculation results, adjust the feature selection of the model;

[0121] When the accuracy is lower than the preset value, the system will automatically adjust the feature selection of the model and automatically select features with a small Gini index.

[0122] It should be specifically noted that: the calculated model accuracy is compared with the preset threshold. If the accuracy is lower than the preset value, the feature selection is automatically adjusted. For features with a large Gini index, their weight in the model is reduced to reduce their impact on the model accuracy.

[0123] Reference Figure 4 As shown, S3 specifically includes:

[0124] Initialize the performance of the device and obtain the initialization performance indicators of the device;

[0125] Obtaining data transmission demand forecast;

[0126] Calculate the load index of the device based on the predicted data transmission demand and the initialization performance of the device;

[0127] The equipment load index calculation formula is: F 负 =GF×F J

[0128] Among them, F 负 is the load index of the equipment, GF is the predicted data transmission demand, and F J is the initialization performance of the Jth device;

[0129] Design a priority calculation model based on all equipment load indicators;

[0130] Based on the priority calculation model, the initial priority of the data is calculated;

[0131] Based on the priority calculation results, the interactive scheduling of data is allocated;

[0132] The priority calculation formula is:

[0133]

[0134] Among them, Y jt is the priority of the jth device at time t, p is the total number of hysteresis learning stages, m is the total number of devices, and α j is the j-th order autoregressive coefficient, F Jtt- is the load forecast of the jth device at time ti;

[0135] ∈ t is the error term of the t-th unit time.

[0136] It should be specifically noted that the performance of the initialization device and the initialization performance indicators of the device include:

[0137] Confirm the hardware specifications of the device, including but not limited to: processor, memory, storage, network interface, graphics card and accelerator card;

[0138] Determine the operating system and software environment that the device will use, including operating system versions, configuration files, software drivers, and applications and tools;

[0139] Assemble and install hardware according to equipment specifications;

[0140] Install the operating system and drivers;

[0141] Conduct performance tests to verify the initialization performance indicators of the equipment;

[0142] Use benchmark tools to measure processor and memory performance;

[0143] Use disk benchmark tools to evaluate storage performance;

[0144] Use a network performance test tool to test network bandwidth and latency.

[0145] Reference Figure 5 As shown, S4 specifically includes:

[0146] Categorize and label data according to identified priority criteria;

[0147] Priority scheduling based on task tags;

[0148] After the scheduling is implemented, the execution process of the task is monitored and feedback is provided;

[0149] Based on the monitoring and feedback results, adjustments and optimizations are made, including re-evaluation of priority criteria.

[0150] What needs to be explained specifically is:

[0151] Based on the results of the analysis, determine whether the current priority standards need to be adjusted. Adjustments include: modifying existing priority levels, adding new priority factors, refining existing factors, and adjusting the priority relationship between different tasks.

[0152] Based on the monitoring and feedback results, adjustments and optimizations are made, including:

[0153] After the scheduling is implemented, during the task execution process, the scheduling progress and resource usage of the task are continuously monitored, and task execution efficiency data feedback is collected. The task execution efficiency includes: execution time, resource utilization and satisfaction;

[0154] Analyze the efficiency of task execution and resource utilization based on the collected monitoring data;

[0155] If monitoring data shows that the actual priority of data tasks does not match expectations, re-evaluate and adjust the priority criteria.

[0156] Monitoring data shows that the actual priority of data tasks is not in line with expectations, so re-evaluate and adjust the priority.

[0157] The specific requirements include:

[0158] Acquire monitoring data of task execution, the monitoring data including task execution time, resource consumption, and order of task completion;

[0159] Determine whether the currently used priority criteria are assigned in order;

[0160] Adjust resource allocation strategies based on the re-evaluated priority criteria based on monitoring results.

[0161] According to the monitoring results and the re-evaluated priority criteria, the resource allocation strategy is adjusted, specifically including: manually updating the priority field in the scheduling system to complete the reallocation of the priority division strategy;

[0162] The scheduling and monitoring process is restarted based on the adjusted priority criteria.

[0163] Implement updated prioritization criteria and continually monitor performance of tasks.

[0164] The use process of the present invention is:

[0165] Step 1: Get data from the API interface and send the data to the cloud platform;

[0166] Step 2: Identify and process missing values ​​in the data through filling methods, detect and remove duplicate data records in the data set, and normalize the data;

[0167] Step 3: Obtain preprocessed data, classify and label the preprocessed data, and identify the data task;

[0168] Step 4: Based on the data task indicators, establish a data demand prediction model to predict data demand;

[0169] Step 5: Initialize the performance of the device;

[0170] Step 6: Calculate the load index of the device based on the predicted data transmission demand and the initialization performance of the device;

[0171] Step 7: Design a priority calculation model based on all equipment load indicators;

[0172] Step 8: Calculate the initial priority of the data based on the priority calculation model;

[0173] Step 9: Based on the priority calculation results, classify and mark the data according to the determined priority criteria;

[0174] Step 10: Priority scheduling based on task tags;

[0175] Step 11: After the scheduling is implemented, monitor and provide feedback on the execution process of the task;

[0176] Step 12: Make adjustments and optimizations based on monitoring and feedback results.

[0177] To sum up, the advantages of the present invention are: it can predict the data transmission requirements in real time, which makes the priority division more convenient and accurate, resulting in a higher accuracy rate of interactive scheduling of data, and is less likely to cause scheduling errors, bringing better usage prospects and commercial value.

[0178] The above shows and describes the basic principles, main features and advantages of the present invention. Those skilled in the art should understand that the present invention is not limited to the above embodiments. The above embodiments and descriptions only describe the principles of the present invention. The present invention may be subject to various changes and improvements without departing from the spirit and scope of the present invention. These changes and improvements fall within the scope of the present invention. The scope of protection claimed by the present invention is defined by the attached claims and their equivalents.

Claims

1. A data interaction scheduling platform based on cloud computing, It is characterized in that The data interaction and scheduling platform includes: S1: Collect historical data and real-time data and pre-process the data; S2: Based on the preprocessed data, obtain the data task, calculate the data transmission demand based on the task indicators, and predict the data transmission demand; S3: Based on the data prediction results, calculate the current device load and prioritize the data; S4: Based on the priority division, data can be interactively scheduled.

2. The data interactive scheduling platform based on cloud computing according to claim 1, It is characterized in that Collect historical data and real-time data, and pre-process the data, including: Obtain API interface and collect historical operation data of cloud computing platform; Identify and process missing values ​​in historical operation data through filling methods, detect and remove duplicate data records in the data set, and normalize the data; The formula for data normalization is: in, is the normalized value, g is the value of the original data, ming is the minimum value in the data set g, and maxg is the maximum value in the data set g.

3. The data interactive scheduling platform based on cloud computing according to claim 1, It is characterized in that S2 specifically includes: Based on the preprocessed historical operation data, classify and mark the preprocessed data to identify the data tasks; based on the data task indicators, predict the data demand; Assume that there are several data tasks that need to be analyzed and supported; For each task, key factors are defined, including: data volume requirements, data update frequency, data quality requirements and data packet loss rate; Build a data demand forecasting model based on the key factors of each task; The data demand prediction model formula is: GF=(β 0 +DQ 1 ×β 1 +DQ 2 ×β 2 +DQ i ×β n ) Among them, DQ i is the i-th key factor parameter, n is the total number of key factors, β 0 , β 1 , β 2 , β n are regression coefficients, and GF represents the predicted data transmission demand.

4. The data interactive scheduling platform based on cloud computing according to claim 3, It is characterized in that Obtain preprocessed data, classify and label the preprocessed data, and identify the data tasks specifically including: Based on the preprocessed historical operation data, task preference marking is performed to obtain the operation data preference classification. Get the running task set; Construct the original data set based on the running task set and divide it into training set and validation set; Based on the preferred feature data and task label data in the training set, the corresponding decision tree model is trained to perform task classification; The decision tree model formula is: Where D is the Gini index of the data set, K is the total number of categories, PK is the proportion of samples belonging to category K in set D, is the predicted category of s; Select the feature with the smallest Gini index as the split node; Use the accuracy evaluation formula to evaluate the model performance, and adjust the feature selection of the model based on the accuracy results.

5. The data interactive scheduling platform based on cloud computing according to claim 4, It is characterized in that Use the accuracy evaluation formula to evaluate the model performance. According to the accuracy results, adjust the feature selection of the model, including: Based on the calculation results of the decision tree model, calculate the accuracy of the model; The accuracy evaluation formula is: Among them, A is the accuracy, CP is the number of correct predictions, and N is the total number of samples; Based on the accuracy calculation results, adjust the feature selection of the model; When the accuracy is lower than the preset value, the system will automatically adjust the feature selection of the model and automatically select features with a small Gini index.

6. The data interactive scheduling platform based on cloud computing according to claim 1, It is characterized in that S3 specifically includes: Initialize the performance of the device and obtain the initialization performance indicators of the device; Obtaining data transmission demand forecast; Calculate the load index of the device based on the predicted data transmission demand and the initialization performance of the device; The equipment load index calculation formula is: F 负 =GF×F J Among them, F 负 is the load index of the equipment, GF is the predicted data transmission demand, and F J is the initialization performance of the Jth device; Design a priority calculation model based on all equipment load indicators; Based on the priority calculation model, the initial priority of the data is calculated; Based on the priority calculation results, the interactive scheduling of data is allocated; The priority calculation formula is: Among them, Y jt is the priority of the jth device at time t, p is the total number of hysteresis learning stages, m is the total number of devices, and α j is the j-th order autoregressive coefficient, F Jt-i is the load forecast of the jth device at time ti; ∈ t is the error term of the t-th unit time.

7. The data interactive scheduling platform based on cloud computing according to claim 1, It is characterized in that S4 specifically includes: Categorize and label data according to identified priority criteria; Priority scheduling based on task tags; After the scheduling is implemented, the execution process of the task is monitored and feedback is provided; Based on the monitoring and feedback results, adjustments and optimizations are made, including re-evaluation of priority criteria.

8. The data interactive scheduling platform based on cloud computing according to claim 7, It is characterized in that Based on the monitoring and feedback results, adjustments and optimizations are made, including: After the scheduling is implemented, during the task execution process, the scheduling progress and resource usage of the task are continuously monitored, and task execution efficiency data feedback is collected. The task execution efficiency includes: execution time, resource utilization and satisfaction; Analyze the efficiency of task execution and resource utilization based on the collected monitoring data; If monitoring data shows that the actual priority of data tasks does not match expectations, re-evaluate and adjust the priority criteria.

9. The data interactive scheduling platform based on cloud computing according to claim 8, It is characterized in that Monitoring data shows that the actual priority of data tasks does not match expectations. The criteria for re-evaluating and adjusting the priority include: Acquire monitoring data of task execution, the monitoring data including task execution time, resource consumption, and order of task completion; Determine whether the currently used priority criteria are assigned in order; Adjust resource allocation strategies based on the re-evaluated priority criteria based on monitoring results.

10. The data interactive scheduling platform based on cloud computing according to claim 9, It is characterized in that According to the monitoring results and the re-evaluated priority criteria, the resource allocation strategy is adjusted, specifically including: manually updating the priority field in the scheduling system to complete the reallocation of the priority division strategy; The scheduling and monitoring process is restarted based on the adjusted priority criteria. Implement updated prioritization criteria and continually monitor performance of tasks.

Citation Information

Cited By

  • AI-based cloud data intelligent analysis and service management system and method

    CN120658632A

  • An ai-based cloud data intelligent analysis and service management system and method

    CN120658632B