Computer information monitoring method and system based on Internet of Things

By monitoring resource occupation and energy consumption in real time in computer equipment groups and reasonably allocating task requests, the problem of unbalanced hardware resource occupation is solved, and task processing efficiency and resource utilization are improved.

CN120256099AActive Publication Date: 2025-07-04JIANGSU VOCATIONAL COLLEGE OF BUSINESS +1
View PDF 8 Cites 0 Cited by

Patent Information

Application Number
CN202510315268.2
Authority / Receiving Office
CN · China
Patent Type
Applications(China)
Current Assignee / Owner
Filing Date
2025-03-18
Publication Date
2025-07-04
Estimated Expiration
2045-03-18

AI Technical Summary

Technical Problem

In the prior art, there is an unbalanced hardware resource occupancy problem in the computer equipment group, resulting in low task processing efficiency and waste of resources. Especially when small enterprises or research centers are difficult to pay high equipment costs, direct queue allocation leads to excessive use of one hardware and other hardware is idle.

Method used

By establishing a group of computer equipment, based on the proportional needs of management equipment and execution equipment, we can monitor resource occupation and energy consumption temperature in real time, generate scheduling tasks and trigger masking management signals, reasonably allocate task requests, prioritize and low-risk equipment management to avoid excessive resource utilization.

Benefits of technology

It realizes more reasonable scheduling of computer resources, avoids excessive hardware use and resource waste caused by unreasonable resource occupation of the same type, and improves task processing efficiency and equipment resource utilization.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN120256099A_ABST
    Figure CN120256099A_ABST
Patent Text Reader

Abstract

The invention relates to the field of computer monitoring and management, and discloses a computer information monitoring method and system based on the Internet of Things, which are used for state monitoring and scheduling management of computer equipment, and are particularly suitable for monitoring and scheduling management of a computer group constructed by networking a plurality of small computer equipment. By monitoring the callable resource devices of the multiple independent devices in the group, the occupation conditions of different types of available resources of the callable resource devices are judged in real time, so that more reasonable task scheduling based on computer resource occupation deviation in the task request processing process is achieved, and the task scheduling efficiency can be effectively improved in the device group. In the embodiment of the invention, the situation that part of types of resources of equipment are occupied less and part of types of resources exceed the limit due to the fact that a plurality of same-type resource occupation tasks are allocated to the same equipment due to unreasonable scheduling is avoided, so that the task scheduling in the group is more reasonable, and the utilization of the equipment resources is more sufficient.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] The present invention relates to the field of computer monitoring and management, and specifically to a computer information monitoring method and system based on the Internet of Things. Background Art

[0002] When a computer device executes a task, the involved device resources include a variety of types, including CPU, GPU, memory, storage, and communication bandwidth, etc. Among them, the communication bandwidth includes the data exchange bandwidth between different device resources and the bandwidth for data interaction with external devices. For different task types, their demands and occupations for different device resources are also different.

[0003] For some small enterprises, research centers and other site objects, their demands for task operations are periodic, and it may be difficult to afford the high procurement cost of professional data center equipment. Therefore, they can only process tasks by using consumer-grade computer devices, or by building a device group with multiple consumer-grade computer devices and processing demands through the way of centralized computing power resources. However, in the prior art, most of them adopt direct queuing for task allocation, which may result in the situation that a single hardware in a certain device in the queue is over-occupied while other hardwares are in an idle state, causing tasks to wait in the queue and affecting the progress of task processing. Summary of the Invention

[0004] The purpose of the present invention is to provide a computer information monitoring method and system based on the Internet of Things to solve the problems raised in the above background art.

[0005] To achieve the above purpose, the present invention provides the following technical solutions:

[0006] A computer information monitoring method based on the Internet of Things, comprising:

[0007] Establish a computer device group, and set a management device and an execution device based on management requirements, and link the management device and the execution device to generate a task scheduling group, where the management requirements are used to represent the proportion requirements of the management device and the execution device;

[0008] Monitor the status of multiple computer devices in the task scheduling group based on the management device, and obtain and update device status data in real time. The device status data includes resource occupancy data and energy consumption temperature data;

[0009] Perform request response through the management device, establish a task list based on task requests, and several task requests in the task list all include the resource occupancy requirements of the corresponding tasks;

[0010] Match the device status data of several devices in the task scheduling group based on resource occupancy requirements, select devices to generate corresponding scheduling tasks, where the scheduling tasks are used to represent the allocation of task requests to the corresponding devices and guide the execution;

[0011] When generating a scheduling task, a shielding management signal is triggered simultaneously. Based on the shielding management signal, the scheduling interface of the current device is shielded for a specified duration, and the specified duration of the shielding management signal is inversely proportional to the task reception volume per unit time of the current management device.

[0012] As a further solution of the present invention: Multiple devices in the task scheduling group all include multiple resource type objects, and the resource occupancy data represents the occupancy of multiple resource type objects at the current time node;

[0013] The energy consumption and temperature data is used to represent the hardware temperature data of the resource type object at the current time node. When the hardware temperature reaches the preset rated temperature threshold, the call priority of the current device hardware is set to the lowest priority. When at the lowest priority, the device hardware is not subject to the priority scheduling management via the resource occupancy data;

[0014] The resource occupancy requirement is used to represent the occupancy requirement of the current task request for multiple resource type objects. Different task requests have different priority sequences for the occupancy preference of different resource type objects based on the difference in the operation logic when they are processed by computer devices.

[0015] As a further solution of the present invention: The step of matching the device status data of several devices in the task scheduling group based on resource occupancy requirements, selecting devices to generate corresponding scheduling tasks specifically includes:

[0016] Sort the devices in the task scheduling group based on the device status data for resource type objects, and obtain the device priority sequence under multiple resource type objects that is updated in real time;

[0017] Sequentially select a specified number of devices in the device priority sequence of the same resource type object, and perform secondary priority sorting on them based on other multiple resource type objects to obtain multiple device sub-priority sequences;

[0018] Obtain the demand priorities and corresponding resource demand quantities of multiple resource type objects in the current resource occupancy requirement, and sequentially match the device priority sequence and the device sub-priority sequence, select the corresponding computer device to establish a scheduling task. If a certain resource in the multiple resource type objects of the device is fully occupied, then the device is marked as the lowest priority.

[0019] As a further aspect of the present invention: when generating a scheduling task, a shielding management signal is simultaneously triggered, and the step of shielding the scheduling interface of the current device for a specified duration based on the shielding management signal specifically includes:

[0020] When generating a scheduling task, trigger a management shielding signal and mark the scheduling task;

[0021] When the corresponding device in the task scheduling group responds to the scheduling task, the management device shields the scheduling interface of the device through the management shielding signal, making its priority at the lowest state value;

[0022] At the end of the specified duration, update the device status data of the computer device, calculate the occupancy of the scheduling task for multiple resource type objects of the device, so as to generate the resource occupancy requirements corresponding to the task request.

[0023] As a further aspect of the present invention: it further includes a standby low-power management step:

[0024] In the task scheduling group, randomly select an execution device and set it as a low-power risk device, and the low-power risk device is used to execute the corresponding task request without resource occupancy requirements;

[0025] When the low-power risk device correspondingly responds to the scheduling task, perform an assignment evaluation based on multiple priority sequences, select the device with the lowest total occupancy of resource type objects to update as the low-power risk device. At this time, this device no longer accepts the corresponding task requests with resource occupancy requirements.

[0026] The embodiment of the present invention aims to provide an Internet of Things-based computer information monitoring system, including:

[0027] A group establishment module, used to establish a computer device group, set a management device and an execution device based on management requirements, and link the management device and the execution device to generate a task scheduling group, where the management requirements are used to represent the occupancy ratio requirements of the management device and the execution device;

[0028] A status synchronization module, used to monitor the status of multiple computer devices in the task scheduling group based on the management device, and obtain and update the device status data in real time, where the device status data includes resource occupancy data and energy consumption temperature data;

[0029] A request response module, used to perform request response through the management device, establish a task list based on the task request, and several task requests in the task list all include the resource occupancy requirements of the corresponding task;

[0030] A task scheduling module, which is used to match the device status data of several devices in a task scheduling group based on resource occupancy requirements, select devices to correspondingly generate scheduling tasks, and the scheduling tasks are used to represent allocating task requests to corresponding devices and guiding execution;

[0031] A scheduling delay module, which is used to trigger a shielding management signal simultaneously when generating a scheduling task, and perform an over-time shielding on the scheduling interface of the current device based on the shielding management signal, and the over-time of the shielding management signal is inversely proportional to the task reception volume per unit time of the current management device.

[0032] As a further solution of the present invention: Multiple devices in the task scheduling group all include multiple resource type objects, and the resource occupancy data represents the occupancy amounts of multiple resource type objects at the current time node;

[0033] The energy consumption temperature data is used to represent the hardware temperature data of the resource type object at the current time node. When the hardware temperature reaches a preset rated temperature threshold, the call priority of the current device hardware is set to the lowest priority. When in the lowest priority, the device hardware is not subject to priority scheduling management via the resource occupancy data;

[0034] The resource occupancy requirement is used to represent the occupancy requirements of the current task request for multiple resource type objects. Different task requests have priority sequence differences in the occupancy bias for different resource type objects based on the arithmetic logic differences when they are processed by a computer device.

[0035] As a further solution of the present invention: The task scheduling module specifically includes:

[0036] A resource sequence unit, which is used to sort the resource type objects of multiple devices in the task scheduling group based on the device status data, and obtain the device priority sequence under multiple resource type objects with real-time updates;

[0037] A secondary sequence unit, which is used to sequentially select a rated number of devices in the device priority sequence of the same resource type object, and perform secondary priority sorting on them based on other multiple resource type objects to obtain multiple device sub-priority sequences;

[0038] A device matching unit, which is used to obtain the demand priorities and corresponding resource demand quantities of multiple resource type objects in the current resource occupancy requirement, and sequentially match the device priority sequence and the device sub-priority sequence, select the corresponding computer device to establish a scheduling task. If a certain resource in the multiple resource type objects of the device is fully occupied, the device is marked as the lowest priority.

[0039] As a further solution of the present invention: The scheduling delay module includes:

[0040] A delay trigger unit, configured to trigger a management shielding signal and mark the scheduling task when a scheduling task is generated;

[0041] A priority shielding unit, configured to, when the corresponding device in the task scheduling group responds to the scheduling task, the management device shields the scheduling interface of the device through the management shielding signal, so that its priority is at the lowest state value;

[0042] A delay synchronization unit, configured to update the device status data of the computer device when the specified duration ends, calculate the occupancy of the scheduling task for multiple resource type objects of the device, so as to generate the resource occupancy requirement corresponding to the task request.

[0043] As a further aspect of the present invention: it further includes a low-power management module, specifically including:

[0044] A low-power setting unit, configured to randomly select an execution device in the task scheduling group and set it as a low-power risk device, and the low-power risk device is used to execute the corresponding task request without resource occupancy requirements;

[0045] A low-power update unit, configured to, when the low-power risk device responds to the scheduling task correspondingly, perform an assignment evaluation based on multiple priority sequences, select the device with the lowest total occupancy of resource type objects to update as the low-power risk device. At this time, the device no longer accepts the corresponding task request with resource occupancy requirements.

[0046] Compared with the prior art, the beneficial effects of the present invention are: it is used for the status monitoring and scheduling management of computer devices, especially applicable to the monitoring and scheduling management of a computer group formed by networking multiple small computer devices. By monitoring the callable resource devices of multiple independent devices in the group, it can judge the occupancy of different types of available resources in real time, so as to realize more reasonable task scheduling based on the computer resource occupancy bias in the task request processing process, and can effectively avoid the situation that in such a device group, due to unreasonable scheduling, multiple tasks with the same type of resource occupancy are assigned to the same device, resulting in low occupancy of some types of resources of the device and exceeding the limit of some types of resources, making the task scheduling in the group more reasonable and the utilization of device resources more sufficient. BRIEF DESCRIPTION OF THE DRAWINGS

[0047] Figure 1 It is a flowchart of a computer information monitoring method based on the Internet of Things.

[0048] Figure 2 It is a flowchart of generating a scheduling task in a computer information monitoring method based on the Internet of Things.

[0049] Figure 3It is a block diagram of the composition of a computer information monitoring system based on the Internet of Things. Specific implementation manner

[0050] In order to make the purpose, technical solution and advantages of the present invention clearer, the present invention will be further described in detail below with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain the present invention and are not used to limit the present invention.

[0051] The following describes in detail the specific implementation manner of the present invention with reference to specific embodiments.

[0052] As Figure 1 described, a computer information monitoring method based on the Internet of Things provided by an embodiment of the present invention includes the following steps:

[0053] S10. Establish a computer device group, set a management device and an execution device based on management requirements, and link the management device and the execution device to generate a task scheduling group, where the management requirements are used to represent the proportion requirements of the management device and the execution device;

[0054] S20. Monitor the status of multiple computer devices in the task scheduling group based on the management device, and obtain and update the device status data in real time. The device status data includes resource occupancy data and energy consumption temperature data;

[0055] S30. Perform request response through the management device, establish a task list based on task requests, and several of the task requests in the task list include the resource occupancy requirements of the corresponding tasks;

[0056] S40. Match the device status data of several devices in the task scheduling group based on the resource occupancy requirements, and select devices to generate corresponding scheduling tasks. The scheduling tasks are used to represent allocating task requests to the corresponding devices and guiding the execution;

[0057] S50. When a scheduling task is generated, a shielding management signal is triggered at the same time, and the scheduling interface of the current device is shielded for a specified duration based on the shielding management signal. The specified duration of the shielding management signal is inversely proportional to the number of task receptions per unit time of the current management device.

[0058] As another preferred embodiment of the present invention, a computer information monitoring method based on the Internet of Things is provided, which is used for the status monitoring and scheduling management of computer devices, and is particularly applicable to the monitoring and scheduling management of a computer group built by networking multiple small computer devices. By monitoring the callable resource devices of multiple independent devices in the group, the occupancy of different types of available resources is judged in real time, so as to realize more reasonable task scheduling based on the preference of computer resource occupancy in the process of task request processing, and effectively avoid the situation that in such a device group, due to unreasonable scheduling, multiple tasks with the same type of resource occupancy are assigned to the same device, resulting in low occupancy of some types of resources of the device and exceeding the limit of some types of resources, making the task scheduling in the group more reasonable and the utilization of device resources more sufficient; in a computer device, when it performs task execution processing, the device resources involved include a variety of types, including CPU, GPU, memory, storage, and communication bandwidth, etc. Among them, the communication bandwidth includes the data exchange bandwidth between different device resources and the bandwidth of data interaction with external devices, etc. For different types of tasks, their requirements and occupancies for different device resources are also different. Therefore, through reasonable task allocation, the computer group can work better and execute tasks, and complete the task content with the highest efficiency; in the prior art, there are professional large-scale task computing centers that can efficiently process tasks. However, for some small enterprises, research centers and other site objects, their demand for task computing is periodic, and it may be difficult to afford the high equipment procurement cost. Therefore, they can only process tasks by using consumer-grade computer devices, or build a device group through multiple consumer-grade computer devices and process the demand by concentrating computing power resources. However, in the prior art, most of them use a direct queue for task allocation, which will result in the situation that a single hardware in a certain device in the queue is occupied too much while other hardware is in an idle state, resulting in tasks waiting in the queue and affecting the progress of task processing. Moreover, when tasks are transferred between devices in the group, it will additionally increase the occupancy of transmission bandwidth resources, causing unnecessary resource reuse waste; the implementation method of this embodiment is: group several computers, and allocate task scheduling groups according to the ratio of corresponding management devices and execution devices. In each scheduling group, the management device monitors the occupancy of various types of resources of the execution device. According to the occupancy of multiple hardware resources of each device, the priority level of the available resources of the device in subsequent task processing can be judged. For the received task requests, their corresponding resource occupancy requirements are pre-obtained (for example, whether the task is a logical processing type task or a graphics processing type task, etc., and based on the basic code and task content, judge the possible occupancy of different types of resources), so that the devices in the current different occupancy states can be matched according to the requirements, and the task requests can be assigned to the corresponding devices, effectively avoiding the problem of lag in task processing efficiency caused by excessive hardware occupancy.

[0059] As another preferred embodiment of the present invention, multiple devices in the task scheduling group each include multiple resource type objects, and the resource occupancy data represents the occupancy of multiple resource type objects at the current time node;

[0060] The energy consumption temperature data is used to represent the hardware temperature data of the resource type object at the current time node. When the hardware temperature reaches a preset rated temperature threshold, the call priority of the current device hardware is set to the lowest priority. When at the lowest priority, the device hardware is not subject to the priority scheduling management via the resource occupancy data;

[0061] The resource occupancy requirement is used to represent the occupancy requirement of the current task request for multiple resource type objects. Different task requests have a priority sequence difference in the occupancy preference for different resource type objects based on the operation logic difference when they are processed by the computer device.

[0062] In this embodiment, further explanation is made. The resource type objects include logical processors, graphics processors, memory, etc. Among them, the logical processor also includes independent occupancy monitoring of multiple threads, which can be used to distinguish and judge the specific single-thread frequency requirements and multi-thread cooperation requirements of logical processing tasks, that is, corresponding task allocation; when an abnormal situation occurs in the device, although the resources may not be fully occupied, the heat generated by its power consumption gradually accumulates, indicating that there may be a fault in the hardware or the basic system software. Therefore, the corresponding device may not be able to handle subsequent task requests. Therefore, the energy consumption temperature data can be used for safety verification and monitoring of the device.

[0063] As Figure 2 shown, as another preferred embodiment of the present invention, the step of matching the device state data of several devices in the task scheduling group based on the resource occupancy requirement and selecting devices to generate scheduling tasks specifically includes:

[0064] S41, sort the devices in the task scheduling group based on the device state data for the resource type objects, and obtain the device priority sequence under multiple resource type objects with real-time updates;

[0065] S42, sequentially select a rated number of devices in the device priority sequence of the same resource type object, and perform secondary priority sorting on them based on other multiple resource type objects to obtain multiple device sub-priority sequences;

[0066] S43. Obtain the demand priorities of multiple resource type objects in the current resource occupancy requirements and the corresponding resource demand quantities, and sequentially match the device priority sequence and the device sub-priority sequence, select the corresponding computer device to establish a scheduling task. If a certain resource in the multiple resource type objects of the device is fully occupied, then the device is marked as the lowest priority.

[0067] In this embodiment, the steps for generating a scheduling task are described here. The specific implementation method is a sequential screening process. First, according to the current device monitoring, the specific device status data of multiple devices in the task scheduling group can be clearly obtained, including the occupancy of multiple resource type objects. Therefore, sorting can be performed according to the occupancy of different resource type objects (here specifically ascending order sorting, that is, the higher the idle ratio of the device, the higher the corresponding priority). For each task request, its occupancy priority and demand for different resource type objects are also certain. Therefore, the multiple resource type objects can be sequentially screened and the satisfaction evaluation of the remaining quantity can be performed to select a suitable device to execute the task request.

[0068] As another preferred embodiment of the present invention, when generating a scheduling task, simultaneously trigger a shielding management signal. The steps for performing an elapsed time shielding on the scheduling interface of the current device based on the shielding management signal specifically include:

[0069] When generating a scheduling task, trigger a management shielding signal and mark the scheduling task;

[0070] When the corresponding device in the task scheduling group responds to the scheduling task, the management device shields the scheduling interface of the device through the management shielding signal to make its priority in the lowest state value;

[0071] At the end of the elapsed time, update the device status data of the computer device, calculate the occupancy of the scheduling task for multiple resource type objects of the device, so as to generate the resource occupancy requirements corresponding to the task request.

[0072] In this embodiment, in the foregoing multiple embodiments, during the process of task request scheduling matching, the specific demand of a task request for multiple resource type objects is only a rough reference value (before the task request is fully executed and evaluated). Therefore, it is impossible to accurately judge the subsequent specific occupancy of the device during task allocation. Under inaccurate evaluation, if task reallocation is performed at this time, the newly allocated task and the current task may jointly cause the actual hardware resource demand of the device to exceed the maximum value, resulting in a decrease in task processing efficiency and affecting the work progress. Therefore, after each device obtains a task request, it is necessary to limit task allocation through a short-term management shielding signal, giving the device system a time to execute the currently allocated task and fully run it into the processing process, so as to realize the specific monitoring of the occupancy and ensure the accuracy of subsequent allocation.

[0073] As another preferred embodiment of the present invention, it further includes a standby low-power management step:

[0074] In the task scheduling group, a device to be executed is randomly selected and set as a low-power risk device, and the low-power risk device is used to execute the corresponding task request without resource occupancy requirements;

[0075] When the low-power risk device responds to the scheduling task correspondingly, assignment evaluation is performed based on multiple priority sequences, and the device with the lowest total occupancy of resource type objects is selected to be updated as the low-power risk device. At this time, this device no longer accepts the corresponding task request with resource occupancy requirements.

[0076] In this embodiment, in the task request, there may also be a part that does not contain any information related to resource occupancy requirements at all. Based on the solutions of the foregoing multiple embodiments, such information cannot be effectively scheduled and allocated. Therefore, a low-power risk device is set here. By retaining and idling the devices with low occupancy among multiple computer devices, a more generous resource allocation can be realized, ensuring the uncontrollable resource occupancy of this type of task demand after allocation and scheduling, and maintaining the steady execution management of subsequent task requests.

[0077] As Figure 3 shown, the present invention also provides an Internet of Things-based computer information monitoring system, which includes:

[0078] A group establishment module 100, configured to establish a computer device group, set a management device and an execution device based on management requirements, and link the management device and the execution device to generate a task scheduling group, where the management requirements are used to represent the occupancy ratio requirements of the management device and the execution device;

[0079] The status synchronization module 200 is used to monitor the status of multiple computer devices in the task scheduling group based on the management device, and obtain and update the device status data in real time. The device status data includes resource occupancy data and energy consumption and temperature data;

[0080] The request response module 300 is used to perform request response through the management device and establish a task list based on the task request. Several of the task requests in the task list each contain the resource occupancy requirements for the corresponding task;

[0081] The task scheduling module 400 is used to match the device status data of several devices in the task scheduling group based on the resource occupancy requirements, select devices to generate corresponding scheduling tasks, and the scheduling tasks are used to represent allocating task requests to the corresponding devices and guiding the execution;

[0082] The scheduling delay module 500 is used to, when generating a scheduling task, simultaneously trigger a shielding management signal, and perform an over-time shielding on the scheduling interface of the current device based on the shielding management signal. The over-time of the shielding management signal is inversely proportional to the task reception volume per unit time of the current management device.

[0083] As another preferred embodiment of the present invention, multiple devices in the task scheduling group each include multiple resource type objects, and the resource occupancy data represents the occupancy of multiple resource type objects at the current time node;

[0084] The energy consumption and temperature data is used to represent the hardware temperature data of the resource type object at the current time node. When the hardware temperature reaches the preset rated temperature threshold, the call priority of the current device hardware is set to the lowest priority. When in the lowest priority, the device hardware is not subject to the priority scheduling management via the resource occupancy data;

[0085] The resource occupancy requirements are used to represent the occupancy requirements of the current task request for multiple resource type objects. Different task requests have a priority sequence difference in the occupancy bias for different resource type objects based on the arithmetic logic difference when being processed by the computer device.

[0086] As another preferred embodiment of the present invention, the task scheduling module specifically includes:

[0087] The resource sequence unit is used to sort the resource type objects of multiple devices in the task scheduling group based on the device status data, and obtain the device priority sequence under multiple resource type objects updated in real time;

[0088] The secondary sequence unit is used to sequentially select a rated number of devices in the device priority sequence of the same resource type object, and perform secondary priority sorting on them based on other multiple resource type objects to obtain multiple device sub-priority sequences;

[0089] The device matching unit is used to obtain the demand priorities and corresponding resource demand quantities of multiple resource type objects in the current resource occupancy demand, and sequentially match the device priority sequence and the device sub-priority sequence, select the corresponding computer device to establish a scheduling task. If a certain resource in the multiple resource type objects of the device is fully occupied, then the device is marked as the lowest priority.

[0090] As another preferred embodiment of the present invention, the scheduling delay module includes:

[0091] The delay trigger unit is used to trigger a management shielding signal and mark the scheduling task when a scheduling task is generated;

[0092] The priority shielding unit is used to, when the corresponding device in the task scheduling group responds to the scheduling task, the management device shields the scheduling interface of the device through the management shielding signal to make its priority in the lowest state value;

[0093] The delay synchronization unit is used to update the device status data of the computer device at the end of the specified duration, calculate the occupancy of the scheduling task for multiple resource type objects of the device, so as to generate the resource occupancy demand corresponding to the task request.

[0094] As another preferred embodiment of the present invention, it further includes a low-power management module, specifically including:

[0095] The low-power setting unit is used to randomly select an execution device in the task scheduling group and set it as a low-power risk device, and the low-power risk device is used to execute the corresponding task request without resource occupancy demand;

[0096] The low-power update unit is used to, when the low-power risk device correspondingly responds to the scheduling task, perform an assignment evaluation based on multiple priority sequences, select the device with the lowest total occupancy of resource type objects to update as the low-power risk device. At this time, the device no longer accepts the corresponding task request with resource occupancy demand.

[0097] Those of ordinary skill in the art can understand that all or part of the processes in the methods of the above embodiments can be completed by instructing relevant hardware through a computer program. The program can be stored in a non-volatile computer-readable storage medium. When the program is executed, it can include the processes of the embodiments of the above methods. Among them, any reference to a memory, storage, database, or other medium used in the embodiments provided in the present application can include non-volatile and / or volatile memories. Non-volatile memory can include read-only memory (ROM), programmable ROM (PROM), electrically programmable ROM (EPROM), electrically erasable programmable ROM (EEPROM), or flash memory. Volatile memory can include random access memory (RAM) or external cache memory. By way of illustration and not limitation, RAM is available in various forms, such as static RAM (SRAM), dynamic RAM (DRAM), synchronous DRAM (SDRAM), double data rate SDRAM (DDR SDRAM), enhanced SDRAM (ESDRAM), synchronous link DRAM (SLDRAM), Rambus direct RAM (RDRAM), direct memory bus dynamic RAM (DRDRAM), and Rambus dynamic RAM (RDRAM), etc.

[0098] After considering the specification and the disclosure of the embodiments, those skilled in the art will readily conceive of other embodiments of the present disclosure. The present application is intended to cover any variations, uses, or adaptations of the present disclosure that follow the general principles of the present disclosure and include known common knowledge or conventional technical means in the technical field not disclosed in the present disclosure. The specification and the embodiments are only regarded as exemplary, and the true scope and spirit of the present disclosure are pointed out by the claims.

[0099] It should be understood that the present disclosure is not limited to the exact structures described above and shown in the drawings, and various modifications and changes can be made without departing from its scope. The scope of the present disclosure is only limited by the appended claims.

Claims

1. A computer information monitoring method based on the Internet of Things, characterized in that, Including: Establish a group of computer devices, set a management device and an execution device based on management requirements, and link the management device and the execution device to generate a task scheduling group, where the management requirements are used to represent the proportion requirements of the management device and the execution device; Based on the management device, monitor the status of multiple computer devices in the task scheduling group, and obtain and update device status data in real time. The device status data includes resource occupancy data and energy consumption temperature data; Through the management device for request response, establish a task list based on task requests. Several of the task requests in the task list include the resource occupancy requirements for the corresponding tasks; Based on the resource occupancy requirements, match the device status data of several devices in the task scheduling group, select devices to generate scheduling tasks accordingly. The scheduling tasks are used to represent the allocation of task requests to the corresponding devices and guide the execution; When generating a scheduling task, simultaneously trigger a shielding management signal, and perform an timed shielding on the scheduling interface of the current device based on the shielding management signal. The timed duration of the shielding management signal is inversely proportional to the task reception volume per unit time of the current management device.

2. The computer information monitoring method based on the Internet of Things according to claim 1, characterized in that, Multiple devices in the task scheduling group each include multiple resource type objects, and the resource occupancy data represents the occupancy of multiple resource type objects at the current time node; The energy consumption temperature data is used to represent the hardware temperature data of the resource type object at the current time node. When the hardware temperature reaches a preset rated temperature threshold, the call priority of the current device hardware is set to the lowest priority. When at the lowest priority, the device hardware is not subject to priority scheduling management via the resource occupancy data; The resource occupancy requirements are used to represent the occupancy requirements of the current task request for multiple resource type objects. Different task requests have priority sequence differences in the occupancy bias for different resource type objects based on the arithmetic logic differences when they are processed by computer devices.

3. The computer information monitoring method based on the Internet of Things according to claim 2, wherein, The step of matching the device status data of several devices in the task scheduling group based on the resource occupancy requirements and selecting devices to generate scheduling tasks accordingly specifically includes: Based on the device status data, sort the resource type objects of multiple devices in the task scheduling group to obtain the device priority sequence under multiple resource type objects updated in real time; Sequentially select a rated number of devices in the device priority sequence of the same resource type object, and perform a secondary priority sorting on them based on other multiple resource type objects to obtain multiple device sub-priority sequences; Obtain the demand priorities and corresponding resource demand quantities of multiple resource type objects in the current resource occupancy requirements, and sequentially match the device priority sequence and the device sub-priority sequence, select the corresponding computer device to establish a scheduling task. If a certain resource of the device is fully occupied among multiple resource type objects, then the device is marked as the lowest priority.

4. A computer information monitoring method based on the Internet of Things according to claim 3, characterized in that, The step of when generating a scheduling task, simultaneously trigger a shielding management signal, and perform an timed shielding on the scheduling interface of the current device based on the shielding management signal specifically includes: When generating a scheduling task, trigger a management shielding signal and mark the scheduling task; When the corresponding device in the task scheduling group responds to the scheduling task, the management device shields the scheduling interface of the device through a management shielding signal, making its priority at the lowest state value; At the end of the specified duration, update the device status data of the computer device, and calculate the occupancy of the scheduling task for multiple resource type objects of the device to generate the resource occupancy requirements for the corresponding task request.

5. A computer information monitoring method based on the Internet of Things according to claim 1, characterized in that, It also includes an alternative low-power management step: In the task scheduling group, randomly select an execution device and set it as a low-power risk device, and the low-power risk device is used to execute the corresponding task request without resource occupancy requirements; When the low-power risk device responds to the scheduling task correspondingly, perform an assignment evaluation based on multiple priority sequences, and select the device with the lowest total occupancy of resource type objects to update it as the low-power risk device. At this time, the device no longer accepts the corresponding task requests with resource occupancy requirements.

6. An Internet of Things-based computer information monitoring system, characterized in that, It includes: A group establishment module, which is used to establish a computer device group, set a management device and an execution device based on management requirements, and link the management device and the execution device to generate a task scheduling group, and the management requirements are used to represent the occupancy ratio requirements of the management device and the execution device; A status synchronization module, which is used to monitor the status of multiple computer devices in the task scheduling group based on the management device, and obtain and update the device status data in real time. The device status data includes resource occupancy data and energy consumption temperature data; A request response module, which is used to perform request response through the management device, and establish a task list based on the task request. Several task requests in the task list all include the resource occupancy requirements of the corresponding task; A task scheduling module, which is used to match the device status data of several devices in the task scheduling group based on the resource occupancy requirements, select a device to generate a corresponding scheduling task, and the scheduling task is used to represent the allocation of task requests to the corresponding device and guide the execution; A scheduling delay module, which is used to trigger a shielding management signal when generating a scheduling task, and perform a specified-duration shielding on the scheduling interface of the current device based on the shielding management signal. The specified duration of the shielding management signal is inversely proportional to the task reception volume per unit time of the current management device.

7. An Internet of Things-based computer information monitoring system according to claim 6, characterized in that, Multiple devices in the task scheduling group each include multiple resource type objects, and the resource occupancy data represents the occupancy of multiple resource type objects at the current time node; The energy consumption temperature data is used to represent the hardware temperature data of the resource type object at the current time node. When the hardware temperature reaches the preset rated temperature threshold, the call priority of the current device hardware is set to the lowest priority. When at the lowest priority, the device hardware is not subject to priority scheduling management via the resource occupancy data; The resource occupancy requirement is used to represent the occupancy requirement of the current task request for multiple resource type objects. Different task requests have priority sequence differences in the occupancy bias for different resource type objects due to the difference in the operation logic when being processed by the computer device.

8. An Internet of Things-based computer information monitoring system according to claim 7, characterized in that The task scheduling module specifically includes: A resource sequence unit is used to sort the resource type objects of multiple devices in a task scheduling group based on device status data, and obtain the device priority sequence under multiple real-time updated resource type objects; A secondary sequence unit is used to sequentially select a rated number of devices in the device priority sequence of the same resource type object, and perform secondary priority sorting on them based on other multiple resource type objects to obtain multiple device sub-priority sequences; A device matching unit is used to obtain the demand priorities of multiple resource type objects and the corresponding resource demand quantities in the current resource occupancy requirement, and sequentially match the device priority sequence and the device sub-priority sequence, select the corresponding computer device to establish a scheduling task. If a certain resource in the multiple resource type objects of the device is fully occupied, then the device is marked as the lowest priority.

9. An Internet of Things-based computer information monitoring system according to claim 8, characterized in that, The scheduling delay module includes: A delay trigger unit is used to trigger a management shielding signal and mark the scheduling task when the scheduling task is generated; A priority shielding unit is used to manage the device to shield the scheduling interface of the device through the management shielding signal when the corresponding device in the task scheduling group responds to the scheduling task, so that its priority is in the lowest state value; A delay synchronization unit is used to update the device status data of the computer device at the end of the specified duration, calculate the occupancy of the scheduling task for multiple resource type objects of the device, and generate the resource occupancy requirement corresponding to the task request.

10. An Internet of Things-based computer information monitoring system according to claim 6, characterized in that, It also includes a low-power management module, specifically including: A low-power setting unit is used to randomly select an execution device in the task scheduling group and set it as a low-power risk device. The low-power risk device is used to execute the corresponding task request without resource occupancy requirements; A low-power update unit is used to perform an assignment evaluation based on multiple priority sequences when the low-power risk device responds to the scheduling task correspondingly, select the device with the lowest total occupancy of resource type objects to update as the low-power risk device. At this time, the device no longer accepts the corresponding task request with resource occupancy requirements.

Citation Information

Patent Citations

  • Task scheduling method and device

    CN116775233A

  • Multi-task scheduling management system under NUMA architecture

    CN116820750A

  • GPU cluster service management system and scheduling method

    CN118312324A

  • Heterogeneous multi-core collaborative management and automatic operation and maintenance system

    CN119003138A

  • Task scheduling method and device based on equipment snapshot polling and storage medium

    CN119025287A