Computing power scheduling optimization method and system based on health degree of end-side equipment

Abstract

The invention particularly relates to a computing power scheduling optimization method and system based on the health degree of end-side equipment. According to the computing power scheduling optimization method based on the health degree of the end-side equipment, health degree index data of the equipment is defined and collected in real time, the health degree index data is input into a health degree evaluation model after being preprocessed, and a comprehensive health degree score of each piece of end-side equipment is calculated; a dynamic computing power scheduling strategy is made based on the health degree evaluation result, and key tasks or tasks with high resource requirements are preferentially scheduled to equipment with high health degree to be executed, so that the reliability and efficiency of task execution are ensured. According to the computing power scheduling optimization method and system based on the health degree of the end-side equipment, the key task is preferentially scheduled to the equipment with the high health degree, so that the resource utilization rate and the overall performance are improved, the equipment with performance degradation is prevented from being excessively used, the risk of task failure and system crash caused by the performance degradation of the equipment is reduced, and the service life of the equipment is prolonged. And the overall stability of the edge computing system is improved.

Description

Technical Field

[0001] The present invention relates to the field of blockchain technology, and in particular to a computing power scheduling optimization method and system based on the health of terminal-side devices. Background Art

[0002] With the rapid development of IoT technology and the advent of the era of the Internet of Everything, edge computing, as an emerging paradigm in which data processing and computing tasks are moved to the edge of the network, has become a key technology to support application scenarios such as smart cities, industrial automation, and telemedicine. As a core component of the edge computing architecture, the end-side device is responsible for data collection, preliminary processing, and some computing tasks. Its performance and stability directly affect the performance and user experience of the entire edge computing system.

[0003] However, in actual applications, edge devices often face complex and changeable working environments, including but not limited to harsh conditions such as high temperature, high humidity, electromagnetic interference, and the need for continuous high-load operation. Under the combined effect of these factors, the equipment gradually shows a phenomenon of performance degradation, which is manifested in the decline of resource utilization, increase of error rate, and increase of response time. In the long run, it will not only affect the execution efficiency and success rate of a single task, but may also trigger a chain reaction, resulting in the decline of stability of the entire edge computing system, or even paralysis.

[0004] Traditional computing power scheduling methods are often based on static configuration or simple load balancing strategies, ignoring the dynamic performance changes of edge devices in actual operation. This "one-size-fits-all" scheduling method is difficult to adapt to the dynamic fluctuations of device performance, and cannot effectively respond to sudden high load demands or device failures. Therefore, there is an urgent need for a method that can monitor the health status of the device in real time and dynamically adjust the computing power scheduling strategy accordingly, so as to optimize task allocation and improve the overall stability and performance of the edge computing system.

[0005] Based on the above situation, the present invention proposes a computing power scheduling optimization method and system based on the health of terminal-side equipment. Summary of the invention

[0006] In order to overcome the deficiencies of the prior art, the present invention provides a simple and efficient method and system for optimizing computing power scheduling based on the health of end-side devices.

[0007] The present invention is achieved through the following technical solutions:

[0008] A computing power scheduling optimization method based on the health of a terminal device, characterized in that it includes the following steps:

[0009] Step S1: define and collect health indicator data of the device in real time, including but not limited to resource utilization, error rate and running time;

[0010] In step S1, the resource utilization rate includes the total amount and utilization rate of the CPU of the device, and the total amount and utilization rate of the memory;

[0011] The error rate includes the total number of times tasks are scheduled to the device and the success rate, as well as the number of crashes and crash rate when the device runs tasks.

[0012] Step S2: preprocessing the collected data, including cleaning and sorting, and removing outliers and noise;

[0013] Step S3: Establish a health evaluation model, input the preprocessed health indicator data into the health evaluation model, and calculate the comprehensive health score of each end-side device;

[0014] In step S3, the health assessment model uses the support vector data description SVDD algorithm with dynamic weights and performs customized adjustments according to actual needs to accurately reflect the actual performance status of the device.

[0015] Step S4: formulate a dynamic computing power scheduling strategy based on the health assessment results. When a new task is generated, dynamic computing power scheduling is implemented by combining task information with the comprehensive health score of the device. Prioritize scheduling critical tasks or tasks with high resource requirements to execute on devices with high health to ensure the reliability and efficiency of task execution.

[0016] In step S4, the process of implementing dynamic computing power scheduling is as follows:

[0017] Step S4.1: When a new task is generated, a task priority label is given to the device according to the priority of the task;

[0018] Step S4.2, calculate the remaining resource information of the device, and perform preliminary screening according to the priority of the task and the resources required for the task; then, dynamically schedule the task according to the task priority label on the device and the comprehensive health score of the device;

[0019] Step S4.3: The comprehensive health score of the device decreases as the resource utilization rate increases. When the comprehensive health score of the device is lower than the custom threshold, some or all tasks are automatically scheduled to other devices with a comprehensive health score higher than the custom threshold until the comprehensive health score of the device is no lower than the custom threshold.

[0020] Step S5: Set a scheduled task or preset trigger condition, re-evaluate the device health periodically or according to the preset trigger condition, and adaptively adjust the computing power scheduling strategy;

[0021] Adjustments to the computing power scheduling strategy include starting, stopping, redeploying, and deleting tasks.

[0022] A computing power scheduling optimization system based on the health of terminal devices includes:

[0023] The health indicator data definition and collection module is responsible for defining and collecting the health indicator data of the device in real time, including but not limited to resource utilization, error rate and running time;

[0024] The resource utilization includes the total amount and utilization rate of the device's CPU, and the total amount and utilization rate of the device's memory;

[0025] The error rate includes the total number of times and success rate of tasks scheduled to the device, and the number of crashes and crash rate of tasks running on the device.

[0026] The preprocessing module is responsible for preprocessing the collected data, including cleaning and sorting, and removing outliers and noise;

[0027] The health assessment module is responsible for establishing a health assessment model, inputting the pre-processed health indicator data into the health assessment model, and calculating the comprehensive health score of each end-side device;

[0028] The health assessment model adopts the support vector data description SVDD algorithm with dynamic weights and is customized according to actual needs to accurately reflect the actual performance status of the equipment.

[0029] The dynamic computing power scheduling module is responsible for formulating dynamic computing power scheduling strategies based on the health assessment results. When a new task is generated, dynamic computing power scheduling is implemented by combining task information with the comprehensive health score of the device. Key tasks or tasks with high resource requirements are scheduled to be executed on devices with high health first, to ensure the reliability and efficiency of task execution.

[0030] The process of implementing dynamic computing power scheduling by the dynamic computing power scheduling module is as follows:

[0031] Step S4.1: When a new task is generated, a task priority label is given to the device according to the priority of the task;

[0032] Step S4.2, calculate the remaining resource information of the device, and perform preliminary screening according to the priority of the task and the resources required for the task; then, dynamically schedule the task according to the task priority label on the device and the comprehensive health score of the device;

[0033] Step S4.3: The comprehensive health score of the device decreases as the resource utilization rate increases. When the comprehensive health score of the device is lower than the custom threshold, some or all tasks are automatically scheduled to other devices with a comprehensive health score higher than the custom threshold until the comprehensive health score of the device is no lower than the custom threshold.

[0034] The update module is responsible for setting scheduled tasks or preset trigger conditions, re-evaluating the health of the device regularly or according to the preset trigger conditions, and adaptively adjusting the computing power scheduling strategy;

[0035] Adjustments to the computing power scheduling strategy include starting, stopping, redeploying, and deleting tasks.

[0036] A computing power scheduling and optimization device based on the health of an end-side device, characterized in that it includes a memory and a processor; the memory is used to store a computer program, and the processor is used to implement the method described above when executing the computer program.

[0037] A readable storage medium, characterized in that: a computer program is stored on the readable storage medium, and the computer program implements the above method when executed by a processor.

[0038] The beneficial effects of the present invention are as follows: the computing power scheduling optimization method and system based on the health of the terminal device reduces the risk of task failure and system crash caused by device performance degradation by giving priority to scheduling key tasks to devices with high health, thereby improving the overall stability of the edge computing system;

[0039] At the same time, by dynamically adjusting the computing power scheduling strategy, resources can be allocated and utilized more reasonably, improving resource utilization and overall performance, avoiding excessive use of equipment with performance degradation, and helping to extend the service life of end-side equipment and reduce maintenance costs. BRIEF DESCRIPTION OF THE DRAWINGS

[0040] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, the drawings described below are some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.

[0041] Attached Figure 1 Schematic diagram of the computing power scheduling optimization method based on the health of the terminal side device of the present invention. DETAILED DESCRIPTION

[0042] In order to enable those skilled in the art to better understand the technical solutions in the present invention, the technical solutions in the embodiments of the present invention will be clearly and completely described below in conjunction with the embodiments of the present invention. Obviously, the described embodiments are only a part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work should fall within the scope of protection of the present invention.

[0043] The computing power scheduling optimization method based on the health of the terminal device includes the following steps:

[0044] Step S1, define and collect health indicator data of the device in real time, including but not limited to resource utilization (CPU, memory, storage, etc.), error rate (hardware error, software crash, etc.) and running time;

[0045] In step S1, the resource utilization rate includes the total amount and utilization rate of the CPU of the device, and the total amount and utilization rate of the memory;

[0046] The error rate includes the total number of times tasks are scheduled to the device and the success rate, as well as the number of crashes and crash rate when the device runs tasks.

[0047] The above health indicator data are collected in real time through built-in or external sensors and software monitoring tools.

[0048] Step S2: preprocessing the collected data, including cleaning and sorting, removing outliers and noise, and making the data more accurate and reliable;

[0049] Step S3: Establish a health evaluation model, input the preprocessed health indicator data into the health evaluation model, and calculate the comprehensive health score of each end-side device;

[0050] In step S3, the health assessment model uses the support vector data description SVDD algorithm with dynamic weights and performs customized adjustments according to actual needs to accurately reflect the actual performance status of the device.

[0051] Step S4: formulate a dynamic computing power scheduling strategy based on the health assessment results. When a new task is generated, dynamic computing power scheduling is implemented by combining task information with the comprehensive health score of the device. Prioritize scheduling critical tasks or tasks with high resource requirements to execute on devices with high health to ensure the reliability and efficiency of task execution.

[0052] In step S4, the process of implementing dynamic computing power scheduling is as follows:

[0053] Step S4.1: When a new task is generated, a task priority label is given to the device according to the priority of the task;

[0054] Step S4.2, calculate the remaining resource information of the device, and perform preliminary screening according to the priority of the task and the resources required for the task; then, dynamically schedule the task according to the task priority label on the device and the comprehensive health score of the device;

[0055] Step S4.3: The comprehensive health score of the device decreases as the resource utilization rate increases. When the comprehensive health score of the device is lower than the custom threshold, some or all tasks are automatically scheduled to other devices with a comprehensive health score higher than the custom threshold until the comprehensive health score of the device is no lower than the custom threshold.

[0056] Step S5: Considering that the device status may change over time, set a scheduled task or preset trigger condition, re-evaluate the device health periodically or according to the preset trigger condition, and adaptively adjust the computing power scheduling strategy;

[0057] Adjustments to the computing power scheduling strategy include starting, stopping, redeploying, and deleting tasks.

[0058] The computing power scheduling optimization system based on the health of the terminal device includes:

[0059] The health indicator data definition and collection module is responsible for defining and collecting the health indicator data of the device in real time, including but not limited to resource utilization (CPU, memory, storage, etc.), error rate (hardware error, software crash, etc.) and running time;

[0060] The resource utilization includes the total amount and utilization rate of the device's CPU, and the total amount and utilization rate of the device's memory;

[0061] The error rate includes the total number of times and success rate of tasks scheduled to the device, and the number of crashes and crash rate of tasks running on the device.

[0062] The above health indicator data are collected in real time through built-in or external sensors and software monitoring tools.

[0063] The preprocessing module is responsible for preprocessing the collected data, including cleaning and sorting, removing outliers and noise, and making the data more accurate and reliable;

[0064] The health assessment module is responsible for establishing a health assessment model, inputting the pre-processed health indicator data into the health assessment model, and calculating the comprehensive health score of each end-side device;

[0065] The health assessment model adopts the support vector data description SVDD algorithm with dynamic weights and is customized according to actual needs to accurately reflect the actual performance status of the equipment.

[0066] The dynamic computing power scheduling module is responsible for formulating dynamic computing power scheduling strategies based on the health assessment results. When a new task is generated, dynamic computing power scheduling is implemented by combining task information with the comprehensive health score of the device. Key tasks or tasks with high resource requirements are scheduled to be executed on devices with high health first, to ensure the reliability and efficiency of task execution.

[0067] The process of implementing dynamic computing power scheduling by the dynamic computing power scheduling module is as follows:

[0068] Step S4.1: When a new task is generated, a task priority label is given to the device according to the priority of the task;

[0069] Step S4.2, calculate the remaining resource information of the device, and perform preliminary screening according to the priority of the task and the resources required for the task; then, dynamically schedule the task according to the task priority label on the device and the comprehensive health score of the device;

[0070] Step S4.3: The comprehensive health score of the device decreases as the resource utilization rate increases. When the comprehensive health score of the device is lower than the custom threshold, some or all tasks are automatically scheduled to other devices with a comprehensive health score higher than the custom threshold until the comprehensive health score of the device is no lower than the custom threshold.

[0071] The update module is responsible for setting scheduled tasks or preset trigger conditions, re-evaluating the health of the device regularly or according to the preset trigger conditions, and adaptively adjusting the computing power scheduling strategy;

[0072] Adjustments to the computing power scheduling strategy include starting, stopping, redeploying, and deleting tasks.

[0073] The computing power scheduling and optimization device based on the health of the terminal device includes a memory and a processor; the memory is used to store a computer program, and the processor is used to implement the method described above when executing the computer program.

[0074] The readable storage medium stores a computer program, and when the computer program is executed by a processor, the method described above is implemented.

[0075] Compared with the existing technology, the computing power scheduling optimization method and system based on the health of the terminal device has the following characteristics:

[0076] First, by prioritizing the scheduling of critical tasks to devices with high health, the risk of task failure and system crash caused by device performance degradation is reduced, and the overall stability of the edge computing system is improved.

[0077] Second, dynamically adjust the computing power scheduling strategy to make resources more reasonably allocated and utilized, thereby improving resource utilization and overall performance.

[0078] Third, it avoids excessive use of equipment with degraded performance, helps extend the service life of end-side equipment and reduce maintenance costs.

[0079] The embodiment described above is only one specific implementation of the present invention. Common changes and substitutions made by those skilled in the art within the scope of the technical solution of the present invention should be included in the protection scope of the present invention.

Claims

1. A computing power scheduling optimization method based on the health of end-side devices, characterized by: The following steps are involved: Step S1: define and collect health indicator data of the device in real time, including but not limited to resource utilization, error rate and running time; Step S2: preprocessing the collected data, including cleaning and sorting, and removing outliers and noise; Step S3: Establish a health evaluation model, input the preprocessed health indicator data into the health evaluation model, and calculate the comprehensive health score of each end-side device; Step S4: formulate a dynamic computing power scheduling strategy based on the health assessment results. When a new task is generated, dynamic computing power scheduling is implemented by combining task information with the comprehensive health score of the device. Prioritize scheduling critical tasks or tasks with high resource requirements to execute on devices with high health to ensure the reliability and efficiency of task execution. Step S5: Set a scheduled task or preset trigger condition, re-evaluate the device health periodically or according to the preset trigger condition, and adaptively adjust the computing power scheduling strategy; Adjustments to the computing power scheduling strategy include starting, stopping, redeploying, and deleting tasks.

2. The computing power scheduling optimization method based on the health of the terminal device according to claim 1 is characterized in that: In step S1, the resource utilization rate includes the total amount and utilization rate of the CPU of the device, and the total amount and utilization rate of the memory; The error rate includes the total number of times tasks are scheduled to the device and the success rate, as well as the number of crashes and crash rate when the device runs tasks.

3. The computing power scheduling optimization method based on the health of the terminal device according to claim 1 is characterized in that: In step S3, the health assessment model uses the support vector data description SVDD algorithm with dynamic weights and performs customized adjustments according to actual needs to accurately reflect the actual performance status of the device.

4. The computing power scheduling optimization method based on the health of the terminal device according to claim 1 is characterized in that: In step S4, the process of implementing dynamic computing power scheduling is as follows: Step S4.1: When a new task is generated, a task priority label is given to the device according to the priority of the task; Step S4.2, calculate the remaining resource information of the device, and perform preliminary screening according to the priority of the task and the resources required for the task; then, dynamically schedule the task according to the task priority label on the device and the comprehensive health score of the device; Step S4.3: The comprehensive health score of the device decreases as the resource utilization rate increases. When the comprehensive health score of the device is lower than the custom threshold, some or all tasks are automatically scheduled to other devices with a comprehensive health score higher than the custom threshold until the comprehensive health score of the device is no lower than the custom threshold.

5. A computing power scheduling optimization system based on the health of end-side devices, characterized by: include: The health indicator data definition and collection module is responsible for defining and collecting the health indicator data of the device in real time, including but not limited to resource utilization, error rate and running time; The preprocessing module is responsible for preprocessing the collected data, including cleaning and sorting, and removing outliers and noise; The health assessment module is responsible for establishing a health assessment model, inputting the pre-processed health indicator data into the health assessment model, and calculating the comprehensive health score of each end-side device; The dynamic computing power scheduling module is responsible for formulating dynamic computing power scheduling strategies based on health assessment results. When a new task is generated, dynamic computing power scheduling is implemented by combining task information with the comprehensive health score of the device. Key tasks or tasks with high resource requirements are scheduled to be executed on devices with high health first, to ensure the reliability and efficiency of task execution. The update module is responsible for setting scheduled tasks or preset trigger conditions, re-evaluating the health of the device regularly or according to the preset trigger conditions, and adaptively adjusting the computing power scheduling strategy; Adjustments to the computing power scheduling strategy include starting, stopping, redeploying, and deleting tasks.

6. The computing power scheduling optimization system based on the health of the terminal device according to claim 5 is characterized in that: The resource utilization includes the total amount and utilization rate of the device's CPU, and the total amount and utilization rate of the device's memory; The error rate includes the total number of times and success rate of tasks scheduled to the device, and the number of crashes and crash rate of tasks running on the device.

7. The computing power scheduling optimization system based on the health of the terminal device according to claim 5 is characterized in that: The health assessment model adopts the support vector data description SVDD algorithm with dynamic weights and is customized according to actual needs to accurately reflect the actual performance status of the equipment.

8. The computing power scheduling optimization system based on the health of the terminal device according to claim 5 is characterized in that: The process of implementing dynamic computing power scheduling by the dynamic computing power scheduling module is as follows: Step S4.1: When a new task is generated, a task priority label is given to the device according to the priority of the task; Step S4.2, calculate the remaining resource information of the device, and perform preliminary screening according to the priority of the task and the resources required for the task; then, dynamically schedule the task according to the task priority label on the device and the comprehensive health score of the device; Step S4.3: The comprehensive health score of the device decreases as the resource utilization rate increases. When the comprehensive health score of the device is lower than the custom threshold, some or all tasks are automatically scheduled to other devices with a comprehensive health score higher than the custom threshold until the comprehensive health score of the device is no lower than the custom threshold.

9. A computing power scheduling optimization device based on the health of the terminal side device, characterized by: The method comprises a memory and a processor; the memory is used to store a computer program, and the processor is used to implement the method according to any one of claims 1 to 4 when executing the computer program.

10. A readable storage medium, characterized in that: The readable storage medium stores a computer program, and when the computer program is executed by a processor, the method according to any one of claims 1 to 4 is implemented.

Images