A computing power cluster management method, device, equipment and storage medium

Abstract

A computing power cluster management method, device, equipment and storage medium, relating to the field of cloud computing technology. In this method, a target management request sent by a user is received; a preset resource scheduling method is used to process the task to be processed, and multiple resource tasks are obtained; multiple edge computing devices are obtained from a device management platform; a first edge computing device is obtained from multiple edge computing devices; when the first edge computing device meets the preset conditions, the first resource task is allocated to the first edge computing device, so that the first edge computing device performs a corresponding operation according to the first resource task, and the first resource task is any one of the multiple resource tasks. The technical solution provided in this application is implemented, the target resources are processed by a preset resource scheduling method, and computing resources are reasonably allocated and utilized based on task requirements and the characteristics of edge devices, thereby realizing unified management and scheduling of massive heterogeneous edge computing resources.

Description

Technical Field

[0001] The present application relates to the field of cloud computing technology, and in particular to a computing cluster management method, device, equipment and storage medium. Background Art

[0002] The current development trend of cloud computing is mainly concentrated in two core directions. First, following the traditional technical context and guided by resource concentration, it is committed to building large-scale or even ultra-large-scale data centers, and through these centers, IT resources are uniformly allocated to achieve optimal utilization and efficient allocation of resources to meet the growing computing and storage needs. Second, it focuses on the coordination and management of heterogeneous cloud computing resources, especially multi-cloud management has become a representative exploration in this field. Its purpose is to achieve resource interoperability and flexible allocation between different cloud service platforms to adapt to the ever-changing business needs and market environment.

[0003] However, with the rapid development of 5G technology, edge computing, as a frontier field of information technology and communication technology, has received widespread attention. Unlike traditional cloud computing, edge computing focuses on the effective management of peripheral or edge resources to meet the needs of low-latency, high-bandwidth data processing. However, there are a large number of edge computer devices with different computing architectures, and this heterogeneity brings many problems. Differences in computing frameworks may lead to fluctuations in performance, and compatibility issues increase the complexity and difficulty of cross-platform development.

[0004] Therefore, how to achieve unified management and scheduling of massive heterogeneous edge computing resources is an urgent problem that needs to be solved. Summary of the invention

[0005] The present application provides a computing cluster management method, apparatus, device and storage medium. The method processes target resources through a preset resource scheduling method, reasonably allocates and utilizes computing resources based on task requirements and the characteristics of edge devices, and realizes unified management and scheduling of massive heterogeneous edge computing resources.

[0006] In the first aspect, the present application provides a computing power cluster management method, which is applied to a device management platform. The method includes: receiving a target management request sent by a user, the target management request includes a task to be processed; using a preset resource scheduling method to process the task to be processed to obtain multiple resource tasks; obtaining multiple edge computing devices from the device management platform, and obtaining a first edge computing device from multiple edge computing devices; when the first edge computing device meets the preset conditions, allocating the first resource task to the first edge computing device, so that the first edge computing device performs corresponding operations according to the first resource task, the first resource task is any one of the multiple resource tasks, and the preset conditions include load conditions.

[0007] By adopting the above technical solution, the tasks to be processed are processed according to the preset resource scheduling method. Based on the task requirements and the characteristics of the edge devices, computing resources can be reasonably allocated and utilized. Multiple edge computing devices can be obtained from the device management platform, and tasks to be processed can be selected and allocated from multiple edge computing devices. Dynamic allocation and scheduling of tasks to be processed are realized, so that the first edge computing device can perform corresponding operations. The unified management and scheduling solution can ensure the stability of the entire system, and improve resource utilization by centralized management and coordination of the operations of each edge computing device.

[0008] Optionally, historical edge computing devices are obtained, and the historical edge computing devices include a second edge computing device; device report information corresponding to the second edge computing device is obtained, and the device report information includes target performance information and target resource information, the target performance information is functional performance information corresponding to the second edge computing device, and the target resource information is resource information corresponding to the second edge computing device; a device management platform is constructed, and the device management platform includes a correspondence between the second edge computing device and the device report information.

[0009] By adopting the above technical solution, the device report information of the historical edge computing devices is obtained, so that the device management platform can clearly understand the performance characteristics and resource status of each computing device, establish the correspondence between the device report information and the second edge computing device, and store the correspondence in the device management platform. By building a device management platform, a large number of edge computing devices can be easily managed and scheduled, providing strong support for various application scenarios.

[0010] Optionally, when the first edge computing device meets the preset conditions, the first resource task is allocated to the first edge computing device so that the first edge computing device performs the corresponding operation according to the first resource task. The method also includes: obtaining a first memory resource, which is an available memory resource corresponding to the first edge computing device; obtaining a second memory resource, which is a memory resource required to execute the first resource task; when the first memory resource is less than the second memory resource, it is confirmed that the first edge computing device cannot process the first resource task, and a third edge computing device is obtained, which is an alternative device corresponding to the first edge computing device; and the first resource task is migrated to the third edge computing device, so that the third edge computing device performs the corresponding operation according to the first resource task.

[0011] By adopting the above technical solution, the first memory resources of the first edge computing device are compared with the second memory resources required to execute the first resource task, and it is accurately determined whether the first edge computing device meets the execution conditions of the first resource task. When the first memory resources are less than the second memory resources, an alternative device for the first edge computing device, that is, a third edge computing device, is found, thereby ensuring that the task can be executed on a device that meets the resource requirements, improving the adaptability of resources, and ensuring the smooth execution of the first resource task.

[0012] Optionally, before using a preset resource scheduling method to process the task to be processed and obtaining multiple resource tasks, the method also includes: obtaining target requirements, the target management request also includes target requirements, the target requirements include target computing resource requirements, target storage resource requirements and target network bandwidth requirements; obtaining a fourth edge computing device from a preset database, and obtaining the target resource usage of the fourth edge computing device, the target resource usage includes target memory status, target disk status and target network status; when the target requirement matches the target resource usage, the fourth edge computing device is used as the target device, so that the preset resource scheduling method is used on the target device to process the task to be processed.

[0013] By adopting the above technical solution, the target demand is obtained, and at the same time, the target resource usage corresponding to the fourth edge computing device is obtained, and the target demand is matched with the target resource usage to ensure that the task is assigned to the device that can meet other resource requirements, thereby improving the efficiency of task execution. The resource usage of the edge computing device is quickly obtained through the preset database, and the task is assigned immediately after matching the appropriate device, thereby improving the response speed and flexibility of the system.

[0014] Optionally, when the first edge computing device meets the preset conditions, the first resource task is allocated to the first edge computing device so that the first edge computing device performs corresponding operations according to the first resource task, specifically including: obtaining multiple computing devices, and obtaining multiple load conditions corresponding to the multiple computing devices in turn, one computing device corresponding to one load condition; sorting the multiple load conditions from small to large to obtain a load sorting result; obtaining a first load condition, the first load condition is the load condition corresponding to the first position in the load sorting result; obtaining a second load condition, the second load condition is the load condition corresponding to the last position in the load sorting result; allocating the processing tasks according to a preset method to obtain multiple resource tasks, and obtaining the first resource task from the multiple resource tasks; preferentially allocating the first resource task to the computing device corresponding to the first load condition, so that the first edge computing device performs corresponding operations according to the first resource task, and the first edge device is the computing device corresponding to the first load condition.

[0015] By adopting the above technical solution, the load conditions of multiple computing devices are sorted, and the lightest load, i.e., the first load condition, is quickly identified, and the heaviest load, i.e., the second load condition, is identified. Sorting makes resource allocation more efficient, and the first resource task is preferentially allocated to the first edge computing device to ensure that the task is executed on a device with better performance and more sufficient resources, which helps to improve the execution efficiency of resource tasks and further improve the user experience.

[0016] Optionally, when the first edge computing device meets the preset conditions, the first resource task is allocated to the first edge computing device so that the first edge computing device performs corresponding operations according to the first resource task, specifically including: obtaining multiple estimated execution times corresponding to multiple resource tasks, one resource task corresponds to one estimated execution time; sorting the multiple estimated execution times to obtain an execution sorting result; obtaining the target estimated execution time, the target estimated execution time is the estimated execution time corresponding to the first one in the execution sorting result; obtaining the first edge computing device, and scheduling the second resource task to the computing device corresponding to the first load condition, so that the first edge computing device performs corresponding operations according to the second resource task, the second resource task is the task corresponding to the target estimated execution time.

[0017] By adopting the above technical solution, the estimated execution times corresponding to multiple resource tasks are obtained and sorted, the estimated completion time of each resource task is clarified, and the second resource task corresponding to the target estimated execution time is preferentially scheduled to the first edge computing device for execution, ensuring that the second resource task can be completed as soon as possible, which helps to improve task execution efficiency and reduce task waiting time.

[0018] Optionally, before allocating the first resource task to the first edge computing device, the method also includes: obtaining a target identity, which is the identity corresponding to the first edge computing device; determining whether the target identity exists in a preset identity table; if the target identity exists in the preset identity table, confirming that the identity authentication of the first edge computing device is successful, and allocating the first resource task to the first edge computing device.

[0019] By adopting the above technical solution, the target identity corresponding to the first edge computing device is obtained, and it is determined whether the target identity exists in the preset identity table, and the identity of the first edge computing device is verified. When the target identity exists in the preset identity table, it is confirmed that the identity authentication of the first edge computing device is successful, and then the first resource task is allocated, ensuring that the first resource task is only allocated to a verified and trusted device, avoiding problems such as task execution errors or data leakage caused by illegal device identity.

[0020] In a second aspect of the present application, a computing power cluster management device is provided, which is a management platform, and the management platform includes an acquisition unit, a processing unit and an allocation unit; the acquisition unit receives a target management request sent by a user, and the target management request includes a task to be processed; the processing unit uses a preset resource scheduling method to process the task to be processed to obtain multiple resource tasks; multiple edge computing devices are obtained from the device management platform, and a first edge computing device is obtained from the multiple edge computing devices; the allocation unit allocates a first resource task to the first edge computing device when the first edge computing device meets a preset condition, so that the first edge computing device performs a corresponding operation according to the first resource task, the first resource task is any one of the multiple resource tasks, and the preset condition includes a load condition.

[0021] Optionally, the acquisition unit is used to acquire historical edge computing devices, and the historical edge computing devices include a second edge computing device; the acquisition unit is used to acquire device report information corresponding to the second edge computing device, and the device report information includes target performance information and target resource information, the target performance information is functional performance information corresponding to the second edge computing device, and the target resource information is resource information corresponding to the second edge computing device; the processing unit is used to build a device management platform, and the device management platform includes the correspondence between the second edge computing device and the device report information.

[0022] Optionally, the acquisition unit is used to acquire a first memory resource, which is an available memory resource corresponding to the first edge computing device; the acquisition unit is used to acquire a second memory resource, which is a memory resource required to execute the first resource task; the processing unit is used to confirm that the first edge computing device cannot process the first resource task when the first memory resource is smaller than the second memory resource, acquire a third edge computing device, which is an alternative device corresponding to the first edge computing device; and migrate the first resource task to the third edge computing device, so that the third edge computing device performs corresponding operations according to the first resource task.

[0023] Optionally, the acquisition unit is used to obtain target requirements, and the target management request also includes target requirements, and the target requirements include target computing resource requirements, target storage resource requirements, and target network bandwidth requirements; the acquisition unit is used to obtain the fourth edge computing device from a preset database, and obtain the target resource usage of the fourth edge computing device, and the target resource usage includes the target memory status, the target disk status, and the target network status; the processing unit is used to use the fourth edge computing device as the target device when the target requirement matches the target resource usage, so as to process the data to be processed on the target device using a preset resource scheduling method.

[0024] Optionally, the acquisition unit is used to acquire multiple computing devices, and to acquire multiple load conditions corresponding to the multiple computing devices in turn, with one computing device corresponding to one load condition; the processing unit is used to sort the multiple load conditions from small to large to obtain a load sorting result; the acquisition unit is used to acquire a first load condition, and the first load condition is the load condition corresponding to the first position in the load sorting result; the acquisition unit is used to acquire a second load condition, and the second load condition is the load condition corresponding to the last position in the load sorting result; the processing unit is used to allocate the processing tasks in a preset manner, obtain multiple resource tasks, and acquire the first resource task from the multiple resource tasks; the allocation unit is used to preferentially allocate the first resource task to the computing device corresponding to the first load condition, so that the first edge computing device performs the corresponding operation according to the first resource task, and the first edge device is the computing device corresponding to the first load condition.

[0025] Optionally, the acquisition unit is used to obtain multiple estimated execution times corresponding to multiple resource tasks, one resource task corresponds to one estimated execution time; the processing unit is used to sort the multiple estimated execution times to obtain an execution sorting result; the acquisition unit is used to obtain the target estimated execution time, the target estimated execution time is the estimated execution time corresponding to the first position in the execution sorting result; the acquisition unit is used to obtain the first edge computing device, and schedule the second resource task to the first edge computing device, so that the first edge computing device performs corresponding operations according to the second resource task, and the second resource task is the task corresponding to the target estimated execution time.

[0026] Optionally, the acquisition unit is used to obtain a target identity, which is the identity corresponding to the first edge computing device; the processing unit is used to determine whether the target identity exists in a preset identity table; if the target identity exists in the preset identity table, it is confirmed that the identity authentication of the first edge computing device is successful, and the first resource task is allocated to the first edge computing device.

[0027] In a third aspect of the present application, an electronic device is provided, which includes a processor, a memory, a user interface and a network interface, the memory is used to store instructions, the user interface and the network interface are used to communicate with other devices, and the processor is used to execute the instructions stored in the memory, so that an electronic device executes any one of the methods described above in the present application.

[0028] In a fourth aspect of the present application, a computer-readable storage medium is provided, wherein the computer-readable storage medium stores instructions, and when the instructions are executed, any one of the above methods of the present application is executed.

[0029] In summary, one or more technical solutions provided in the embodiments of the present application have at least the following technical effects or advantages:

[0030] 1. Process the tasks to be processed according to the preset resource scheduling method. Based on the task requirements and the characteristics of the edge devices, it can reasonably allocate and utilize computing resources, obtain multiple edge computing devices from the device management platform, select and allocate tasks to be processed from multiple edge computing devices, and realize dynamic allocation and scheduling of tasks to be processed, so that the first edge computing device can perform corresponding operations. The unified management and scheduling solution can ensure the stability of the entire system, and improve resource utilization by centrally managing and coordinating the operations of each edge computing device.

[0031] 2. Compare the first memory resources of the first edge computing device with the second memory resources required to execute the first resource task, accurately determine whether the first edge computing device meets the execution conditions of the first resource task, and when the first memory resources are less than the second memory resources, find an alternative device for the first edge computing device, that is, the third edge computing device, so as to ensure that the task can be executed on a device that meets the resource requirements, improve the adaptability of resources, and ensure the smooth execution of the first resource task.

[0032] 3. Sort the load conditions of multiple computing devices, quickly identify the lightest load, i.e. the first load condition, and identify the heaviest load, i.e. the second load condition. Sorting makes resource allocation more efficient, and prioritizes allocating the first resource task to the first edge computing device to ensure that the task is executed on a device with better performance and more resources, which helps to improve the execution efficiency of resource tasks and further improve user experience. BRIEF DESCRIPTION OF THE DRAWINGS

[0033] Figure 1 It is a flowchart of a computing power cluster management method provided in an embodiment of the present application;

[0034] Figure 2 It is a structural diagram of a computing power cluster management device provided in an embodiment of the present application;

[0035] Figure 3 It is a structural schematic diagram of an electronic device disclosed in an embodiment of the present application.

[0036] Explanation of the reference numerals: 201, acquisition unit; 202, processing unit; 203, allocation unit; 300, electronic device; 301, processor; 302, communication bus; 303, user interface; 304, network interface; 305, memory. DETAILED DESCRIPTION

[0037] In order to enable technicians in this field to better understand the technical solutions in this specification, the technical solutions in the embodiments of this specification will be clearly and completely described below in conjunction with the drawings in the embodiments of this specification. Obviously, the described embodiments are only part of the embodiments of this application, not all of the embodiments.

[0038] In the description of the embodiments of the present application, words such as "for example" or "for example" are used to indicate examples, illustrations or explanations. Any embodiment or design described as "for example" or "for example" in the embodiments of the present application should not be interpreted as being more preferred or more advantageous than other embodiments or designs. Specifically, the use of words such as "for example" or "for example" is intended to present related concepts in a specific way.

[0039] In the description of the embodiments of the present application, the meaning of the term "multiple" refers to two or more. For example, multiple systems refer to two or more systems, and multiple screen terminals refer to two or more screen terminals. In addition, the terms "first" and "second" are used for descriptive purposes only and cannot be understood as indicating or implying relative importance or implicitly indicating the indicated technical features. Thus, the features defined as "first" and "second" may explicitly or implicitly include one or more of the features. The terms "include", "comprise", "have" and their variations all mean "including but not limited to", unless otherwise specifically emphasized.

[0040] The current development trend of cloud computing is mainly concentrated in two core directions. First, following the traditional technical context and guided by resource concentration, it is committed to building large-scale or even ultra-large-scale data centers, and through these centers, IT resources are uniformly allocated to achieve optimal utilization and efficient allocation of resources to meet the growing computing and storage needs. Second, it focuses on the coordination and management of heterogeneous cloud computing resources, especially multi-cloud management has become a representative exploration in this field. Its purpose is to achieve resource interoperability and flexible allocation between different cloud service platforms to adapt to the ever-changing business needs and market environment.

[0041] However, with the rapid development of 5G technology, edge computing, as a frontier field of information technology and communication technology, has received widespread attention. Unlike traditional cloud computing, edge computing focuses on the effective management of peripheral or edge resources to meet the needs of low-latency, high-bandwidth data processing. However, there are a large number of edge computer devices with different computing architectures, and this heterogeneity brings many problems. Differences in computing frameworks may lead to fluctuations in performance, and compatibility issues increase the complexity and difficulty of cross-platform development.

[0042] Therefore, how to achieve unified management and scheduling of massive heterogeneous edge computing resources has become an urgent problem to be solved. A computing power cluster management method provided in an embodiment of the present application is applied to a device management platform. The device management platform of the present application can be a platform for resource scheduling of edge computing devices. Figure 1 This is a flowchart of a computing power cluster management method provided by an embodiment of the present application, refer to Figure 1The method includes the following steps S101-S104.

[0043] S101: receiving a target management request sent by a user, where the target management request includes a task to be processed.

[0044] In the above S101, the user sends a target management request through the platform interface. In this case, the user refers to a person who has resources to be processed. After receiving the target management request sent by the user, the device management platform parses the target management request and identifies the pending tasks that the user needs to schedule. The pending tasks are the data to be processed. After parsing the target management request, the identification management platform verifies the legitimacy and validity of the target management request to ensure that the user has the authority to access the pending tasks.

[0045] S102: Processing the task to be processed using a preset resource scheduling method to obtain a plurality of resource tasks.

[0046] In the above S102, after receiving the target management request, the identification management platform uses the preset resource scheduling method to process the pending task. Before obtaining multiple resource tasks, it is necessary to select a suitable edge computing device in the device management platform so that the edge computing device can formulate a task scheduling strategy according to the target resources. Specifically, it includes: obtaining the target demand, the target management request also includes the target demand, the target demand includes the target computing resource demand, the target storage resource demand and the target network bandwidth demand; obtaining the fourth edge computing device from the preset database, obtaining the target resource usage of the fourth edge computing device, the target resource usage includes the target memory status, the target disk status and the target network status; when the target demand matches the target resource usage, the fourth edge computing device is used as the target device, so that the preset resource scheduling method is used on the target device to process the pending task.

[0047] Specifically, after the device management platform receives the target management request sent by the user, the target management request also includes the target demand, which generally includes the target computing resource demand, the target storage resource demand and the target network bandwidth demand. The device management platform will quantify the target demand and convert it into a specific value or range, such as the required number of CPU cores, memory size, disk capacity and network bandwidth. The fourth edge computing device is obtained from the preset database, which stores multiple edge computing devices and target resource usage. In order to ensure that the resource usage corresponding to each edge computing device in the preset database is consistent with the actual situation, the resource usage of each edge computing device in the preset database can be adjusted by real-time monitoring of the usage of each edge computing device, so that the resource usage of each edge computing device in the preset database is synchronized with the resource usage of the actual edge computing device. After obtaining the fourth edge computing device from the preset database, the target resource usage corresponding to the fourth edge computing device is obtained, and the target resource usage includes memory usage, target disk usage and target network usage. The target resource usage can come from direct feedback from the fourth edge device, or it can be obtained through a special monitoring tool. The target demand is compared with the target resource usage of the fourth edge computing device, including comparing the matching degree of computing resources, the adequacy of storage resources, and the availability of network bandwidth. The target resource usage is compared according to the target demand in turn, and the matching degree is evaluated according to the comparison result. If the target resource usage of the fourth edge computing device meets the target demand, the corresponding matching degree is high, and the fourth edge computing device can be regarded as a target device. Since the target resource usage includes three situations, different scores can be set according to the three situations, and the matching degree is determined by the score; if the target resource usage is the same as the target demand, the score corresponding to the target resource usage is obtained; when the target resource usage is different from the target demand, the score is 0, and the three situations are matched with the three requirements in the target demand in turn, and the obtained scores are added to obtain the total score of the target resource usage being the same as the target demand. The matching degree is then determined according to the total score. Different scores can set different matching degrees. The larger the score, the higher the corresponding matching degree; the smaller the score, the lower the corresponding matching degree. After determining that the fourth edge computing device is the target device, the device management platform determines to send the target resources to the fourth edge computing device (ie, the target device) so as to execute a preset resource scheduling method on the target device to process the tasks to be processed.

[0048] In addition, if the device management platform finds multiple target devices, the final target device can be determined based on some additional selection criteria. In this application, memory and hard disk have different meanings. The hard disk can also be named as a disk. Memory usually refers to random access memory (RAM), which is used to temporarily store data and programs that the computer is using or about to use. The read and write speed of memory is very fast, and it is where the CPU directly exchanges data. The hard disk is a non-volatile storage device used for long-term storage of data and the computer's operating system. Even after the power is turned off, the data on the hard disk will not be lost. The memory is mainly used to perform multitasking and high-speed data exchange, and is a temporary working space for the computer to run programs. The hard disk is used to store large amounts of data, including documents, pictures, videos, programs, etc., and is a long-term storage solution for computers. The speed of memory is much higher than that of the hard disk, especially the interaction speed with the CPU, which is crucial for computing tasks that require fast response. The capacity of memory is usually smaller than that of the hard disk, but the unit cost is higher, so it is not suitable for use as large-capacity storage. Memory is volatile, which means that the data stored in the memory will be lost after the power is turned off. The hard disk is non-volatile, and the data can still be retained even after the power is turned off. Memory and hard disk play complementary roles in a computer. Memory is used for fast data exchange and program running, while hard disk is used for long-term storage of data.

[0049] Furthermore, the pending tasks are processed according to the preset resource scheduling method, which is designed based on factors such as the type, quantity, priority, and geographical location of the task. According to the scheduling strategy, the pending tasks can be split into multiple resource tasks, each of which represents an independent operation or processing process. The split tasks may be sorted, and the tasks can be combined or adjusted according to certain optimization principles.

[0050] S103: Acquire multiple edge computing devices from the device management platform, and acquire a first edge computing device from the multiple edge computing devices.

[0051] In the above S103, before obtaining multiple edge computing devices from the device management platform, it is necessary to build a device management platform, which specifically includes: obtaining historical edge computing devices, including the second edge computing device; obtaining device report information corresponding to the second edge computing device, the device report information includes target performance information and target resource information, the target performance information is the functional performance information corresponding to the second edge computing device, and the target resource information is the resource information corresponding to the second edge computing device; building a device management platform, the device management platform includes the corresponding relationship between the second edge computing device and the device report information. Specifically, first access a preset database, the preset database stores the information of the previously used edge computing devices, and the historical edge computing devices are screened out according to certain query conditions (such as device type, usage time range, etc.). During the query process, further screening is performed according to specific standards or rules, and one or more of the screened devices are selected as the second edge computing device. After selecting the second edge computing device, the device report information corresponding to the second edge computing device is obtained, and the target performance information and target resource information are extracted from the device report information. The target performance information includes data on the processing capability, response time, stability, etc. of the device. The target resource information covers the storage capacity, available bandwidth, remaining power and other resource conditions of the edge computing device. The extracted information is verified and cleaned to ensure the accuracy and completeness of the data, and to detect whether the data is complete and not missing and whether the data conforms to the expected format and scope. According to actual needs, a device management platform architecture is designed. This architecture needs to support the registration, management, monitoring and other functions of the device, and can conveniently display the device report information. The acquired second edge computing device and device report information are entered into the database of the device management platform, and the corresponding relationship between the second edge computing device and the device report information is established to ensure that the information can be accurately called and displayed in the subsequent management and monitoring process. The historical edge computing device and its corresponding device report information can be obtained, and a fully functional device management platform can be built. This will help to better manage and monitor all edge computing devices, improve the utilization efficiency of the equipment and the stability of the system.

[0052] In addition, after the device management platform is built, the device management platform can centrally manage all edge computing devices. Through a unified user interface, API interface and management tools, users can easily manage and configure various heterogeneous devices, and then obtain multiple edge computing devices. At this time, multiple edge computing devices refer to online and available devices.

[0053] Furthermore, a first edge computing device is obtained from the multiple edge computing devices, where the first edge computing device is any computing device among the multiple edge computing devices.

[0054] S104: When the first edge computing device meets a preset condition, a first resource task is allocated to the first edge computing device so that the first edge computing device performs a corresponding operation according to the first resource task, where the first resource task is any one of the multiple resource tasks.

[0055] In the above S104, when the first resource task is allocated to the first edge computing device, it is necessary to confirm that the first edge computing device meets the preset conditions. The preset conditions include load conditions and expected execution time conditions. The load condition represents the minimum load strategy, and the preset execution time condition represents the shortest time strategy. The two allocation conditions are introduced in turn below.

[0056] The first allocation method, when the preset condition is a load condition; obtain multiple computing devices, and sequentially obtain multiple load conditions corresponding to the multiple computing devices, one computing device corresponds to one load condition; sort the multiple load conditions from small to large to obtain a load sorting result; obtain a first load condition, the first load condition is the load condition corresponding to the first position in the load sorting result; obtain a second load condition, the second load condition is the load condition corresponding to the last position in the load sorting result; allocate the target resources according to the preset method, obtain multiple resource tasks, and obtain the first resource task from the multiple resource tasks; preferentially allocate the first resource task to the computing device corresponding to the first load condition, so that the first edge computing device performs the corresponding operation according to the first resource task, and the first edge device is the computing device corresponding to the first load condition. Specifically, multiple computing devices are obtained from the device management platform, and the computing devices obtained at this time are available edge computing devices. For each computing device, the current load condition of the acquirer is obtained, and the load condition includes indicators such as CPU usage, memory occupancy, disk, and network bandwidth. Match each computing device with its corresponding load condition and organize it into a data structure that is easy to process. After determining multiple computing devices, multiple load conditions corresponding to the multiple computing devices are obtained in sequence, and one computing device corresponds to one load condition. The sorting standard can be determined according to actual needs. It is usually desired to give priority to assigning resource tasks to computing devices with lighter loads. Multiple load conditions can be sorted in order from small to large to obtain a load sorting result. The computer device corresponding to the first position in the sorting result receives the device with the lightest load, and the computing device corresponding to the last position is the device with the heaviest load. From the sorting result, the device management platform obtains the load condition corresponding to the first position, that is, the first load condition, which represents the lightest computing device. Similarly, the load condition corresponding to the last position is obtained from the sorting result, that is, the second load condition, which represents the computing device with the heaviest load. The target resources can be allocated according to a preset method, which includes function and quantity, that is, division according to the number of multiple computing devices, or division according to different functions of the computing devices, and then the target resources are divided into multiple resource tasks, each of which may require different computing resources and storage resources. From multiple resource tasks, obtain the first resource task, determine that the first resource task is the most urgent task among the current target resources, and preferentially assign the first resource task to the computing device corresponding to the first load condition. Since the computing device corresponding to the first load condition has the lightest current load, it can better handle new tasks, thereby ensuring the efficiency and stability of the execution of the first resource task.

[0057] The second allocation method, the preset condition is the estimated execution time condition. In addition to determining to allocate the first resource task to the first edge computing device according to the load situation, the allocation can also be made according to the execution time of each resource task, specifically including: obtaining multiple estimated execution times corresponding to multiple resource tasks, one resource task corresponds to one estimated execution time; sorting the multiple estimated execution times to obtain the execution sorting result; obtaining the target estimated execution time, the target estimated execution time is the estimated execution time corresponding to the first one in the execution sorting result; obtaining the computing device corresponding to the first load situation, and scheduling the second resource task to the first edge computing device, so that the first edge computing device performs the corresponding operation according to the second resource task, and the second resource task is the task corresponding to the target estimated execution time. Specifically, multiple estimated execution times corresponding to multiple resource tasks are obtained, one resource task corresponds to one estimated execution time, and multiple resource tasks are obtained by dividing the target resources. For each resource task, an estimated execution time is calculated based on its computing requirements, data size, computing device performance and other factors. This time is an estimated value used to represent the time required for the task to be completed under ideal conditions. Match each resource task with its corresponding estimated execution time and integrate them into an easy-to-process data structure. The multiple estimated execution times can be sorted according to actual needs, and the tasks with shorter estimated execution times are usually executed first to maximize the utilization rate of the computing device and the error rate of the tasks. The sorting includes sorting from small to large or from large to small. The multiple estimated execution times can be sorted in the order from small to large to obtain the execution sorting result. The resource task corresponding to the first place in the execution sorting result is the task with the shortest estimated execution time. From the execution sorting result, the estimated execution time corresponding to the first place is obtained, that is, the target estimated execution time. The target estimated execution time represents the execution time corresponding to the resource task with the shortest estimated execution time among all the tasks to be assigned. According to the previous load sorting result, the computing device with the lightest load is determined, that is, the computing device corresponding to the first load situation. The resource task corresponding to the target estimated execution time is found from the multiple resource tasks, that is, the second resource task. The second resource task is the task with the shortest estimated execution time among all the resource tasks to be assigned. The second resource task can be scheduled to the computing device corresponding to the first load situation. Since the computing device has the lightest load, it can process new resource tasks faster and ensure the efficient execution of resource tasks. After receiving the second resource task, the first edge computing device performs corresponding operations according to the requirements of the second resource task. At this time, the second resource task represents the first resource task.

[0058] In addition, after using any of the above allocation methods to determine that the first resource task is allocated to the first edge computing device, when the first resource task is allocated to the first edge computing device, it can be allocated through network transmission. When the first resource task is allocated to the first edge computing device, it is allocated based on the matching degree between the first resource task and the first edge computing device, that is, the first edge computing device can process the first resource task. After the first edge computing device receives the first resource task, it performs corresponding operations according to the first volunteer task. The device management platform monitors the execution process of the first resource task to ensure that the first resource task can be completed correctly and efficiently.

[0059] Furthermore, when the device management platform determines to allocate the first resource task to the first edge computing device, it is necessary to authenticate the identity of the first edge computing device to ensure that only authenticated devices can participate in the execution of the resource task. Specifically, it includes: obtaining the target identity, which is the identity corresponding to the first edge computing device; judging whether the target identity exists in the preset identity table; if the target identity exists in the preset identity table, it is confirmed that the identity authentication of the first edge computing device is successful, and the first resource task is allocated to the first edge computing device.

[0060] Specifically, when the first resource task is retrieved to the first edge computing device, the detailed information of the first edge computing device is first obtained, and the identity identifier is extracted from the first edge computing device. The identity identifier is unique and is used to distinguish different devices. Before allocating resources, the device management platform needs to determine all edge computing devices participating in resource allocation. Then, the identity identifiers of all edge computing devices are obtained, and a table is established to store multiple identity identifiers. The table is named a preset identity identifier table. The preset identity identifier table stores all verified and authorized device identity identifiers. The target identity identifier is compared with the identifier recorded in the preset identity identifier table to find out whether there is a match. When there is a record matching the target identity identifier in the preset identity identifier table, it is confirmed that the identity authentication of the first edge computing device is successful and can participate in the execution of the first resource task. After confirming that the identity authentication is successful, the first resource task is assigned to the first edge computing device. When the target identity identifier does not exist in the preset identity identifier table, it will be tasked that the identity authentication of the first edge computing device fails, and some security measures can be taken, such as refusing to allocate resource tasks and sending alarms, to ensure the security and stability of the device management platform.

[0061] In one possible implementation, when the first edge computing device does not meet the preset conditions, indicating that the memory resources of the first edge computing device are insufficient to execute the first resource task, the first resource task can be migrated to other edge computing devices for continued execution, specifically including: obtaining a first memory resource, the first memory resource is the available memory resource corresponding to the first edge computing device; obtaining a second memory resource, the second memory resource is the memory resource required to execute the first resource task; when the first memory resource is less than the second memory resource, it is confirmed that the first edge computing device cannot process the first resource task, obtaining a third edge computing device, the third edge computing device is an alternative device corresponding to the first edge computing device; and migrating the first resource task to the third edge computing device, so that the third edge computing device performs corresponding operations according to the first resource task.

[0062] Specifically, the memory resource information of the first edge computing device is queried, and the amount of memory resources currently available to the first edge computing device, i.e., the first memory resource, is calculated according to the query result. The first memory resource generally considers factors such as the total memory, used memory, and reserved memory of the device. When analyzing the first resource task, the amount of memory resources required in the process of executing the first resource task is determined, which can be estimated based on factors such as the type, complexity, and data size of the task. Based on the task analysis result, the amount of memory resources required to execute the first resource task, i.e., the second memory resource, is determined. Comparing the size of the first memory resource and the second memory resource is a key step in determining whether the first edge computing device has enough memory to execute the first resource task. If the first memory resource is smaller than the second memory resource, it is confirmed that the first edge computing device cannot effectively perform the first resource task, because insufficient memory may cause the task to execute slowly, error, or fail to complete at all. Multiple memory resources corresponding to other devices are obtained from the device management platform, and the multiple memory resources include a third memory resource, and then the third memory resource is compared with the second memory resource. If the third memory resource is larger than the second memory resource, the edge computing device corresponding to the third memory resource, i.e., the third edge computing device, is obtained. The third edge computing device can be set as an alternative device for the first edge computing device. The third edge computing device is an alternative device prepared in advance for the first edge computing device when the first edge computing device cannot handle the task. The alternative device can be based on various factors such as the performance, geographical location, and load conditions of the device. Before migrating the first resource task, it is necessary to save the current state of the first resource task, including processed data, intermediate results, etc., to ensure that the first resource task can continue to execute seamlessly after migration. Migrating the first resource task and its related status information to the third edge computing device usually involves network transmission and task status synchronization. After the migration is completed, the third edge computing device continues to perform operations according to the requirements of the first resource task, and the device management platform will monitor the execution process of the first resource task to ensure that the migrated first resource task can be successfully completed.

[0063] For example, if the target resource is image data to be processed, and the first resource task is to process high-definition images, after the first resource task is assigned to the first edge computing device, the available memory resources of the first edge computing device, i.e., the first memory resource, are obtained. The first memory resource is set to 4GB. The first resource task is evaluated to determine that the second memory resource is 8GB. At this time, the first memory resource is smaller than the second memory resource, and it is confirmed that the first edge computing device cannot process the first resource task. Then, the third memory resource corresponding to other devices is obtained from the device management platform. If the third memory resource is 16GB, the third memory resource is larger than the second memory resource. The third edge computing device corresponding to the third memory resource is obtained, and the third edge computing device is set as an alternative device for the first edge computing device. Finally, the first resource task is migrated to the third edge computing device so that the third edge computing device can process the first resource task.

[0064] Furthermore, when the first memory resource is greater than or equal to the second memory resource, it is confirmed that the first edge computing device can process the first resource task, and there is no need to migrate the first resource task to other edge computing devices for processing.

[0065] The present application also provides a computing power cluster management device. Figure 2 This is a schematic diagram of the structure of a computing power cluster management device provided in an embodiment of the present application, with reference to Figure 2 The device is a management platform, which includes an acquisition unit 201, a processing unit 202 and an allocation unit 203.

[0066] The acquisition unit 201 receives a target management request sent by a user, where the target management request includes a task to be processed.

[0067] The processing unit 202 uses a preset resource scheduling method to process the task to be processed to obtain multiple resource tasks; obtains multiple edge computing devices from the device management platform, and obtains a first edge computing device from the multiple edge computing devices.

[0068] The allocation unit 203 allocates the first resource task to the first edge computing device when the first edge computing device meets the preset conditions, so that the first edge computing device performs corresponding operations according to the first resource task. The first resource task is any one of multiple resource tasks, and the preset conditions include load conditions.

[0069] In one possible implementation, the acquisition unit 201 is used to acquire historical edge computing devices, which include a second edge computing device; the acquisition unit 201 is used to acquire device report information corresponding to the second edge computing device, which includes target performance information and target resource information, the target performance information is the functional performance information corresponding to the second edge computing device, and the target resource information is the resource information corresponding to the second edge computing device; the processing unit 202 is used to build a device management platform, which includes a correspondence between the second edge computing device and the device report information.

[0070] In one possible implementation, the acquisition unit 201 is used to acquire a first memory resource, which is an available memory resource corresponding to the first edge computing device; the acquisition unit 201 is used to acquire a second memory resource, which is a memory resource required to execute a first resource task; the processing unit 202 is used to confirm that the first edge computing device cannot process the first resource task when the first memory resource is less than the second memory resource, acquire a third edge computing device, which is an alternative device corresponding to the first edge computing device; and migrate the first resource task to the third edge computing device, so that the third edge computing device performs corresponding operations according to the first resource task.

[0071] In one possible implementation, the acquisition unit 201 is used to obtain the target demand, and the target management request also includes the target demand, and the target demand includes the target computing resource demand, the target storage resource demand, and the target network bandwidth demand; the acquisition unit 201 is used to obtain the fourth edge computing device from a preset database, and obtain the target resource usage of the fourth edge computing device, and the target resource usage includes the target memory status, the target disk status, and the target network status; the processing unit 202 is used to use the fourth edge computing device as the target device when the target demand matches the target resource usage, so as to process the pending task on the target device using the preset resource scheduling method.

[0072] In one possible implementation, the acquisition unit 201 is used to acquire multiple computing devices, and sequentially acquire multiple load conditions corresponding to the multiple computing devices, one computing device corresponding to one load condition; the processing unit 202 is used to sort the multiple load conditions from small to large to obtain a load sorting result; the acquisition unit is used to acquire a first load condition, and the first load condition is the load condition corresponding to the first position in the load sorting result; the acquisition unit 201 is used to acquire a second load condition, and the second load condition is the load condition corresponding to the last position in the load sorting result; the processing unit 202 is used to allocate the processing tasks in a preset manner, obtain multiple resource tasks, and acquire the first resource task from the multiple resource tasks; the allocation unit 203 is used to preferentially allocate the first resource task to the computing device corresponding to the first load condition, so that the first edge computing device performs the corresponding operation according to the first resource task, and the first edge device is the computing device corresponding to the first load condition.

[0073] In one possible implementation, the acquisition unit 201 is used to acquire multiple estimated execution times corresponding to multiple resource tasks, one resource task corresponding to one estimated execution time; the processing unit 202 is used to sort the multiple estimated execution times to obtain an execution sorting result; the acquisition unit 201 is used to acquire the target estimated execution time, the target estimated execution time is the estimated execution time corresponding to the first position in the execution sorting result; the acquisition unit 201 is used to acquire the computing device corresponding to the first load condition, and schedule the second resource task to the first edge computing device, so that the first edge computing device performs the corresponding operation according to the second resource task, and the second resource task is the task corresponding to the target estimated execution time.

[0074] In one possible implementation, the acquisition unit 201 is used to obtain a target identity, which is an identity corresponding to a first edge computing device; the processing unit 202 is used to determine whether the target identity exists in a preset identity table; if the target identity exists in the preset identity table, it is confirmed that the identity authentication of the first edge computing device is successful, and the first resource task is allocated to the first edge computing device.

[0075] It should be noted that: when the device provided in the above embodiment realizes its function, only the division of the above functional modules is used as an example. In actual application, the above functions can be assigned to different functional modules as needed, that is, the internal structure of the device is divided into different functional modules to complete all or part of the functions described above. In addition, the device and method embodiments provided in the above embodiment belong to the same concept, and the specific implementation process is detailed in the method embodiment, which will not be repeated here.

[0076] The present application also discloses an electronic device. Figure 3 , Figure 3The electronic device 300 may include: at least one processor 301 , at least one network interface 304 , a user interface 303 , a memory 305 , and at least one communication bus 302 .

[0077] The communication bus 302 is used to realize the connection and communication between these components.

[0078] The user interface 303 may include a display screen (Display) and a camera (Camera). Optionally, the user interface 303 may also include a standard wired interface and a wireless interface.

[0079] The network interface 304 may optionally include a standard wired interface or a wireless interface (such as a WI-FI interface).

[0080] Among them, the processor 301 may include one or more processing cores. The processor 301 uses various interfaces and lines to connect various parts in the entire server, and executes various functions of the server and processes data by running or executing instructions, programs, code sets or instruction sets stored in the memory 305, and calling data stored in the memory 305. Optionally, the processor 301 can be implemented in at least one hardware form of digital signal processing (Digital Signal Processing, DSP), field programmable gate array (Field-Programmable Gate Array, FPGA), and programmable logic array (Programmable Logic Array, PLA). The processor 301 can integrate one or a combination of a central processing unit (Central Processing Unit, CPU), a graphics processing unit (Graphics Processing Unit, GPU) and a modem. Among them, the CPU mainly processes the operating system, user interface and application requests; the GPU is responsible for rendering and drawing the content to be displayed on the display screen; the modem is used to process wireless communications. It can be understood that the above-mentioned modem may not be integrated into the processor 301, and it can be implemented separately through a chip.

[0081] Among them, the memory 305 may include a random access memory (RAM) or a read-only memory (Read-Only Memory). Optionally, the memory 305 includes a non-transitory computer-readable storage medium. The memory 305 can be used to store instructions, programs, codes, code sets or instruction sets. The memory 305 may include a program storage area and a data storage area, wherein the program storage area. Instructions for implementing an operating system, instructions for at least one function (such as a touch function, a sound playback function, an image playback function, etc.), instructions for implementing the above-mentioned various method embodiments, etc. can be stored; the data storage area can store data involved in the above-mentioned various method embodiments, etc. The memory 305 can also be optionally at least one storage device located away from the aforementioned processor 301.

[0082] like Figure 3 As shown, the memory 305 as a computer storage medium may include an operating system, a network communication module, a user interface module, and an application for computing power cluster management.

[0083] exist Figure 3 In the electronic device 300 shown, the user interface 303 is mainly used to provide an input interface for the user and obtain data input by the user; and the processor 301 can be used to call the application for computing power cluster management stored in the memory 305. When executed by one or more processors, the electronic device executes one or more methods described in the above embodiments.

[0084] It should be noted that, for the aforementioned method embodiments, for the sake of simplicity, they are all described as a series of action combinations, but those skilled in the art should be aware that the present application is not limited by the order of the actions described, because according to the present application, certain steps can be performed in other orders or simultaneously. Secondly, those skilled in the art should also be aware that the embodiments described in the specification are all preferred embodiments, and the actions and modules involved are not necessarily required for the present application.

[0085] In the above embodiments, the description of each embodiment has its own emphasis. For parts that are not described in detail in a certain embodiment, reference can be made to the relevant descriptions of other embodiments.

[0086] In the several embodiments provided in the present application, it should be understood that the disclosed devices can be implemented in other ways. For example, the device embodiments described above are only schematic, such as the division of the units, which is only a logical function division. There may be other division methods in actual implementation, such as multiple units or components can be combined or integrated into another system, or some features can be ignored or not executed. Another point is that the mutual coupling or direct coupling or communication connection shown or discussed can be through some service interfaces, and the indirect coupling or communication connection of devices or units can be electrical or other forms.

[0087] The units described as separate components may or may not be physically separated, and the components shown as units may or may not be physical units, that is, they may be located in one place or distributed on multiple network units. Some or all of the units may be selected according to actual needs to achieve the purpose of the solution of this embodiment.

[0088] In addition, each functional unit in each embodiment of the present application may be integrated into one processing unit, or each unit may exist physically separately, or two or more units may be integrated into one unit. The above-mentioned integrated unit may be implemented in the form of hardware or in the form of software functional units.

[0089] If the integrated unit is implemented in the form of a software functional unit and sold or used as an independent product, it can be stored in a computer-readable memory. Based on this understanding, the technical solution of the present application, or the part that contributes to the prior art, or all or part of the technical solution can be embodied in the form of a software product, which is stored in a memory and includes several instructions for a computer device (which can be a personal computer, server or network device, etc.) to execute all or part of the steps of the method described in each embodiment of the present application. The aforementioned memory includes: various media that can store program codes, such as USB flash drives, mobile hard drives, magnetic disks or optical disks.

[0090] The above is only an exemplary embodiment of the present disclosure and cannot be used to limit the scope of the present disclosure. That is, any equivalent changes and modifications made according to the teachings of the present disclosure are still within the scope of the present disclosure. After considering the disclosure of the specification and the truth of practice, it will be easy for those skilled in the art to think of other embodiments of the present disclosure. This application is intended to cover any variation, use or adaptive change of the present disclosure, which follows the general principles of the present disclosure and includes common knowledge or customary technical means in the technical field that are not recorded in the present disclosure.

Claims

1. A computing power cluster management method, characterized in that: Applied in a device management platform, the method includes: Receiving a target management request sent by a user, wherein the target management request includes a task to be processed; The pending tasks are processed using a preset resource scheduling method to obtain a plurality of resource tasks; before the pending tasks are processed using a preset resource scheduling method to obtain a plurality of resource tasks, the method further includes: obtaining a target requirement, the target management request also including the target requirement, the target requirement including a target computing resource requirement, a target storage resource requirement, and a target network bandwidth requirement; obtaining a fourth edge computing device from a preset database, and obtaining a target resource usage of the fourth edge computing device, the target resource usage including a target memory condition, a target disk condition, and a target network condition; when the target requirement matches the target resource usage, the fourth edge computing device is used as a target device, so as to process the pending tasks on the target device using the preset resource scheduling method; Acquire multiple edge computing devices from the device management platform, and acquire a first edge computing device from the multiple edge computing devices; When the first edge computing device meets the preset condition, the first resource task is allocated to the first edge computing device, so that the first edge computing device performs a corresponding operation according to the first resource task, the first resource task is any one of the multiple resource tasks, and the preset condition includes a load condition; when the first edge computing device meets the preset condition, the first resource task is allocated to the first edge computing device, so that the first edge computing device performs a corresponding operation according to the first resource task, specifically including: obtaining multiple computing devices, sequentially obtaining multiple load conditions corresponding to the multiple computing devices, one computing device corresponding to one load condition; sorting the multiple load conditions from small to large to obtain a load sorting result; obtaining a first load condition, the first load condition is the load condition corresponding to the first position in the load sorting result; obtaining a second load condition, the second load condition is the load condition corresponding to the last position in the load sorting result; allocating the tasks to be processed in a preset manner to obtain multiple resource tasks, and obtaining the first resource task from the multiple resource tasks; preferentially allocating the resource task to the computing device corresponding to the first load condition, so that the first edge computing device performs a corresponding operation according to the first resource task, the first edge computing device is the computing device corresponding to the first load condition; When the first edge computing device meets the preset conditions, the first resource task is allocated to the first edge computing device so that the first edge computing device performs corresponding operations according to the first resource task, specifically including: obtaining multiple estimated execution times corresponding to multiple resource tasks, one resource task corresponds to one estimated execution time; sorting the multiple estimated execution times to obtain an execution sorting result; obtaining a target estimated execution time, the target estimated execution time is the estimated execution time corresponding to the first one in the execution sorting result; obtaining the first edge computing device, and scheduling the second resource task to the first edge computing device, so that the first edge computing device performs corresponding operations according to the second resource task, the second resource task is the task corresponding to the target estimated execution time, and the preset conditions also include an estimated execution time condition.

2. The method according to claim 1, characterized in that Before obtaining a plurality of edge computing devices from the device management platform, the device management platform needs to be constructed, specifically including: Acquire a historical edge computing device, wherein the historical edge computing device includes a second edge computing device; Obtain device report information corresponding to the second edge computing device, the device report information including target performance information and target resource information, the target performance information is functional performance information corresponding to the second edge computing device, and the target resource information is resource information corresponding to the second edge computing device; Construct the device management platform, wherein the device management platform includes a corresponding relationship between the second edge computing device and the device report information.

3. The method according to claim 1, characterized in that After the first edge computing device satisfies a preset condition and the first resource task is allocated to the first edge computing device so that the first edge computing device performs a corresponding operation according to the first resource task, the method further includes: Obtain a first memory resource, where the first memory resource is an available memory resource corresponding to the first edge computing device; Acquire a second memory resource, where the second memory resource is a memory resource required to execute the first resource task; When the first memory resource is less than the second memory resource, it is confirmed that the first edge computing device cannot process the first resource task, and a third edge computing device is obtained, where the third edge computing device is an alternative device corresponding to the first edge computing device; And migrate the first resource task to the third edge computing device so that the third edge computing device performs corresponding operations according to the first resource task.

4. The method according to claim 1, characterized in that: Before allocating the first resource task to the first edge computing device, the method further includes: Obtain a target identity, where the target identity is an identity corresponding to the first edge computing device; Determine whether the target identity exists in the preset identity table; If the target identity exists in the preset identity table, it is confirmed that the identity authentication of the first edge computing device is successful, and the first resource task is allocated to the first edge computing device.

5. A computing power cluster management device, characterized in that: The device is a management platform, and the management platform comprises an acquisition unit (201), a processing unit (202) and an allocation unit (203); The acquisition unit (201) receives a target management request sent by a user, wherein the target management request includes a task to be processed; The processing unit (202) processes the to-be-processed task using a preset resource scheduling method to obtain a plurality of resource tasks; Before the preset resource scheduling method is used to process the pending tasks and obtain multiple resource tasks, it also includes: obtaining target requirements, the target management request also includes the target requirements, and the target requirements include target computing resource requirements, target storage resource requirements, and target network bandwidth requirements; obtaining a fourth edge computing device from a preset database, and obtaining target resource usage of the fourth edge computing device, the target resource usage includes target memory conditions, target disk conditions, and target network conditions; when the target requirement matches the target resource usage, the fourth edge computing device is used as a target device, so that the preset resource scheduling method is used on the target device to process the pending tasks; obtaining multiple edge computing devices from a device management platform, and obtaining a first edge computing device from the multiple edge computing devices; The allocation unit (203) allocates a first resource task to the first edge computing device when the first edge computing device meets a preset condition, so that the first edge computing device performs a corresponding operation according to the first resource task, the first resource task being any one of the plurality of resource tasks, the preset condition including a load condition; the allocation unit (203) allocates a first resource task to the first edge computing device when the first edge computing device meets the preset condition, so that the first edge computing device performs a corresponding operation according to the first resource task, specifically comprising: obtaining a plurality of the computing devices, sequentially obtaining a plurality of load conditions corresponding to the plurality of the computing devices, one computing device corresponding to one load condition; sorting the plurality of load conditions from small to large to obtain a load sorting result; obtaining a first load condition, the first load condition being the load condition corresponding to the first position in the load sorting result; obtaining a second load condition, the second load condition being the load condition corresponding to the last position in the load sorting result; allocating the tasks to be processed in a preset manner to obtain a plurality of the resource tasks, and obtaining a plurality of the load conditions from the plurality of the resource tasks. The method comprises the following steps: obtaining a first resource task from a service; preferentially allocating the resource task to a computing device corresponding to the first load condition, so that the first edge computing device performs a corresponding operation according to the first resource task, and the first edge computing device is a computing device corresponding to the first load condition; when the first edge computing device meets a preset condition, allocating the first resource task to the first edge computing device, so that the first edge computing device performs a corresponding operation according to the first resource task, specifically including: obtaining a plurality of estimated execution times corresponding to a plurality of the resource tasks, one resource task corresponding to one estimated execution time; sorting the plurality of estimated execution times to obtain an execution sorting result; obtaining a target estimated execution time, the target estimated execution time being the estimated execution time corresponding to the first position in the execution sorting result; obtaining the first edge computing device, and scheduling a second resource task to the first edge computing device, so that the first edge computing device performs a corresponding operation according to the second resource task, the second resource task being a task corresponding to the target estimated execution time, and the preset condition also includes an estimated execution time condition.

6. An electronic device, characterized in that: The electronic device (300) comprises a processor (301), a memory (305), a user interface (303) and a network interface (304), wherein the memory (305) is used to store instructions, the user interface (303) and the network interface (304) are used to communicate with other devices, and the processor (301) is used to execute the instructions stored in the memory (305) so that the electronic device (300) executes the method according to any one of claims 1 to 4.

7. A computer-readable storage medium, characterized in that: The computer-readable storage medium stores instructions, and when the instructions are executed, the method according to any one of claims 1 to 4 is performed.

Images