Resource management method and system based on cloud edge collaboration

Abstract

The embodiment of the invention provides a resource management method and system based on cloud edge collaboration. According to the method, firstly, an intelligent AI management platform can be constructed to better allocate computing, storage and network resources; and then resource requirements of the static resources and the dynamic resources can be accurately estimated by monitoring calling frequencies of data interfaces of the static resources and the dynamic resources, so that dynamic allocation of the computing resources is realized, enterprises can be helped to better manage and utilize the computing resources, the stability and reliability of the system are improved, the operation cost is reduced, and the economic benefit is increased. And greater value is created for enterprises. Meanwhile, intelligent scheduling of resources can be achieved based on load balancing, tasks can be automatically distributed to nodes with low loads according to the load conditions of all the nodes, and therefore load balancing is achieved, and the overall performance of the system is improved. Through the dynamic resource allocation process, the resource allocation is more reasonable.

Description

Technical Field

[0001] The present invention relates to the field of computer technology, and in particular to a resource management method and system based on cloud-edge collaboration. Background Art

[0002] Cloud-edge collaboration is an indispensable key link in the industrial Internet and a core supporting technology for realizing the digital transformation of the industry. As a new data computing architecture and organizational form, edge computing has expanded the scope of network computing, extending computing from the cloud center to the edge of the network, and providing intelligent services to users nearby.

[0003] At present, computing power is sunk to the edge of the network closer to the source of objects or data, and real-time analysis algorithms are directly run at the edge layer.

[0004] However, in edge computing scenarios, although multiple edge nodes share the traffic load, the computing and storage capabilities of each node are relatively weak compared to the cloud, resulting in unreasonable resource allocation. Summary of the invention

[0005] The embodiments of the present invention provide a resource management method and system based on cloud-edge collaboration, which can make resource allocation more reasonable.

[0006] In a first aspect, an embodiment of the present invention provides a resource management method based on cloud-edge collaboration, the method comprising:

[0007] Building an intelligent AI management platform based on server performance data, network traffic data, and storage usage data, wherein the intelligent AI management platform is used to manage computing, network, and storage resources;

[0008] Determine resource requirements for static resources and dynamic resources, wherein the static resources refer to fixed files stored on the server and the dynamic resources refer to content dynamically generated on the server side;

[0009] Determine the load of at least one node according to the usage of system resources, wherein the usage of system resources includes at least one of: CPU usage, memory occupancy, and network bandwidth;

[0010] Based on the resource demand conditions of the static resources and the dynamic resources and the load conditions of the at least one node, use the intelligent AI management platform to perform resource management;

[0011] Preferably,

[0012] The determining of resource requirements of static resources and dynamic resources includes:

[0013] Determining the static resource demand based on the periodic changes, seasonal changes and impact of emergencies in the static resource usage;

[0014] Determining whether the interface call frequency of the dynamic resource increases;

[0015] When the frequency of interface calls of the dynamic resource increases, increasing the investment of computing resources;

[0016] When the frequency of interface calls of the dynamic resource decreases, reducing the investment in the computing resources;

[0017] Preferably,

[0018] The step of determining the load of at least one node according to the usage of system resources includes:

[0019] Determine whether there is a target node among the at least one node whose load condition is lower than a first preset threshold, wherein the first preset threshold is used to represent a minimum value of the load condition of the at least one node;

[0020] When the load condition of the target node in the at least one node is lower than the first preset threshold, increasing the task allocation amount of the target node;

[0021] Preferably,

[0022] After the resource demand conditions of the static resources and the dynamic resources and the load conditions of the at least one node are used to manage resources using the intelligent AI management platform, the method further includes:

[0023] Classifying the computing resources and generating target computing resources, wherein the target computing resources include at least one of a central processing unit (CPU), a cloud disk, and a network;

[0024] Based on the target computing resource, determine the usage and demand of at least one of the central processing unit CPU, the cloud disk and the network;

[0025] Based on the usage and demand of at least one of the central processing unit CPU, the cloud disk and the network, a dynamic resource allocation visualization VR system is constructed, wherein the dynamic resource allocation visualization VR system is used to intuitively visualize resource allocation;

[0026] Preferably,

[0027] Using sensors deployed on the hardware system to collect operating data of the hardware system in real time, wherein the operating data includes at least one of historical data, real-time operating data, and external environmental factor data of the hardware system;

[0028] Based on the operating data, determining whether the hardware system is in a fatigue state;

[0029] When it is determined that the hardware system is in the fatigue state, adjusting an operating parameter of the hardware system, wherein the operating parameter is used to alleviate the fatigue state of the hardware system;

[0030] Preferably,

[0031] In a second aspect, an embodiment of the present invention provides a resource management system based on cloud-edge collaboration, the system comprising:

[0032] A platform building module, used to build an intelligent AI management platform based on server performance data, network traffic data, and storage usage data, wherein the intelligent AI management platform is used to perform resource management on computing, network, and storage;

[0033] A resource demand determination module, used to determine the resource demand of static resources and dynamic resources, wherein the static resources refer to fixed files stored on the server, and the dynamic resources refer to content dynamically generated on the server side;

[0034] A load determination module, configured to determine the load of at least one node according to the usage of system resources, wherein the usage of system resources includes at least one of: CPU usage, memory occupancy, and network bandwidth;

[0035] A resource management module, configured to perform resource management using the intelligent AI management platform based on the resource demand conditions of the static resources and the dynamic resources and the load conditions of the at least one node;

[0036] Preferably,

[0037] The resource requirement determination module is further used to execute:

[0038] Determining the static resource demand based on the periodic changes, seasonal changes and impact of emergencies in the static resource usage;

[0039] Determining whether the interface call frequency of the dynamic resource increases;

[0040] When the frequency of interface calls of the dynamic resource increases, increasing the investment of computing resources;

[0041] When the frequency of interface calls of the dynamic resource decreases, reducing the investment in the computing resources;

[0042] Preferably,

[0043] The load determination module is further configured to execute:

[0044] Determine whether there is a target node among the at least one node whose load condition is lower than a first preset threshold, wherein the first preset threshold is used to represent a minimum value of the load condition of the at least one node;

[0045] When the load condition of the target node in the at least one node is lower than the first preset threshold, increasing the task allocation amount of the target node;

[0046] Preferably,

[0047] After the resource management, it further includes: a resource visualization module,

[0048] The resource visualization module is used to perform:

[0049] Classifying the computing resources and generating target computing resources, wherein the target computing resources include at least one of a central processing unit (CPU), a cloud disk, and a network;

[0050] Based on the target computing resource, determine the usage and demand of at least one of the central processing unit CPU, the cloud disk and the network;

[0051] Based on the usage and demand of at least one of the central processing unit CPU, the cloud disk and the network, a dynamic resource allocation visualization VR system is constructed, wherein the dynamic resource allocation visualization VR system is used to intuitively visualize resource allocation;

[0052] Preferably,

[0053] Further comprising: a fatigue processing module;

[0054] The fatigue processing module is used to execute:

[0055] Using sensors deployed on the hardware system to collect operating data of the hardware system in real time, wherein the operating data includes at least one of historical data, real-time operating data, and external environmental factor data of the hardware system;

[0056] Based on the operating data, determining whether the hardware system is in a fatigue state;

[0057] When it is determined that the hardware system is in the fatigue state, operating parameters of the hardware system are adjusted, wherein the operating parameters are used to alleviate the fatigue state of the hardware system.

[0058] In a third aspect, an embodiment of the present invention provides a resource management system based on cloud-edge collaboration, including: at least one memory and at least one processor;

[0059] The at least one memory is used to store a machine-readable program;

[0060] The at least one processor is configured to call the machine-readable program to execute any one of the methods described in the first aspect;

[0061] In a fourth aspect, an embodiment of the present invention provides a computer-readable medium, on which computer instructions are stored. When the computer instructions are executed by a processor, the processor executes any one of the methods described in the first aspect.

[0062] The embodiment of the present invention provides a resource management method and system based on cloud-edge collaboration. In the edge computing scenario, the contradiction between the demand for resources of tasks and the scarcity of edge device resources is prominent. The computing and storage capabilities of edge nodes are relatively weak compared to those of the cloud. In addition, edge nodes also include a variety of heterogeneous nodes and are geographically dispersed, and the bandwidth resources of edge networks are relatively scarce. Therefore, the reasonable allocation of limited resources is the key difficulty in achieving low power consumption and low latency in cloud-edge collaborative architecture. In order to solve the above difficulties, an intelligent AI management platform can be first constructed to better allocate computing, storage and network resources. Then, by monitoring the call frequency of the data interface of static resources and dynamic resources, the resource requirements of static resources and dynamic resources can be accurately estimated, and then the dynamic allocation of computing resources can be realized, which can help enterprises better manage and utilize computing resources, improve the stability and reliability of the system, and reduce operating costs at the same time, creating greater value for enterprises. At the same time, intelligent scheduling of resources can be realized based on load balancing. It can automatically allocate tasks to nodes with lower loads according to the load conditions of each node, thereby achieving load balancing and improving the overall performance of the system. Through the above-mentioned dynamic resource allocation process, resource allocation can be made more reasonable. BRIEF DESCRIPTION OF THE DRAWINGS

[0063] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, the drawings described below are some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.

[0064] Figure 1 It is a flow chart of a resource management method based on cloud-edge collaboration provided by one embodiment of the present invention;

[0065] Figure 2 is a flow chart of another resource management method based on cloud-edge collaboration provided by an embodiment of the present invention;

[0066] Figure 3 It is a schematic diagram of a resource management system based on cloud-edge collaboration provided by one embodiment of the present invention;

[0067] Figure 4 is a schematic diagram of another resource management system based on cloud-edge collaboration provided by an embodiment of the present invention;

[0068] Figure 5 It is a schematic diagram of another resource management system based on cloud-edge collaboration provided by an embodiment of the present invention. DETAILED DESCRIPTION

[0069] In order to make the purpose, technical solutions and advantages of the embodiments of the present invention clearer, the technical solutions in the embodiments of the present invention will be clearly and completely described below in conjunction with the drawings in the embodiments of the present invention. Obviously, the described embodiments are part of the embodiments of the present invention, rather than all the embodiments. Based on the embodiments in the present invention, all other embodiments obtained by ordinary technicians in this field without making creative work are within the scope of protection of the present invention.

[0070] like Figure 1 As shown, an embodiment of the present invention provides a resource management method based on cloud-edge collaboration, which may include the following steps:

[0071] Step 101: Building an intelligent AI management platform based on server performance data, network traffic data, and storage usage data, wherein the intelligent AI management platform is used to manage computing, network, and storage resources;

[0072] Step 102: Determine resource requirements for static resources and dynamic resources, wherein static resources refer to fixed files stored on the server, and dynamic resources refer to content dynamically generated on the server side;

[0073] Step 103: Determine the load of at least one node according to the usage of system resources, wherein the usage of system resources includes at least one of: CPU usage, memory occupancy, and network bandwidth;

[0074] Step 104: Based on the resource demand of static resources and dynamic resources and the load of at least one node, use the intelligent AI management platform to manage resources.

[0075] In an embodiment of the present invention, a resource management method based on cloud-edge collaboration is provided. In the edge computing scenario, the contradiction between the demand for resources of tasks and the scarcity of edge device resources is prominent. The computing and storage capabilities of edge nodes are relatively weak compared to those of the cloud. In addition, edge nodes also include a variety of heterogeneous nodes and are geographically dispersed, and the bandwidth resources of edge networks are relatively scarce. Therefore, the reasonable allocation of limited resources is the key difficulty in achieving low power consumption and low latency in cloud-edge collaborative architecture. In order to solve the above difficulties, an intelligent AI management platform can be first constructed to better allocate computing, storage and network resources. Then, by monitoring the call frequency of the data interface of static resources and dynamic resources, the resource requirements of static resources and dynamic resources can be accurately estimated, and then the dynamic allocation of computing resources can be realized, which can help enterprises better manage and utilize computing resources, improve the stability and reliability of the system, and reduce operating costs at the same time, creating greater value for enterprises. At the same time, intelligent scheduling of resources can be realized based on load balancing. It can automatically allocate tasks to nodes with lower loads according to the load conditions of each node, thereby achieving load balancing and improving the overall performance of the system. Through the above-mentioned dynamic resource allocation process, resource allocation can be made more reasonable.

[0076] In order to achieve reasonable allocation of static resources and dynamic resources, in one embodiment of the present invention, the step of determining resource requirements of static resources and dynamic resources in step 102 in the above embodiment includes:

[0077] Determining the static resource demand based on the periodic changes, seasonal changes and impact of emergencies in the static resource usage;

[0078] Determining whether the interface call frequency of the dynamic resource increases;

[0079] When the frequency of interface calls of the dynamic resource increases, increasing the investment of computing resources;

[0080] When the frequency of interface calls of the dynamic resource decreases, the investment in the computing resources is reduced.

[0081] In an embodiment of the present invention, the resource requirements of static resources and dynamic resources can be accurately estimated based on a deep learning algorithm for load balancing. This algorithm can analyze the use of system resources in real time, predict the resource requirements in the future through deep learning technology, and then realize efficient management and allocation of resources. First, for static resources, the deep learning algorithm will use its powerful data processing and learning capabilities to analyze a large amount of historical data to find out the rules of static resource use. This includes modeling the periodic changes, seasonal changes, and the impact of emergencies in resource usage. Through these models, the deep learning algorithm can predict the demand for static resources in the future, so as to prepare and allocate static resources in advance. For dynamic resources, the deep learning algorithm pays more attention to real-time and dynamics. It monitors the call frequency of the data interface in real time and dynamically adjusts the allocation of computing resources according to the change of the call frequency. When the call frequency increases, the algorithm will automatically increase the input of computing resources to meet the system processing requirements; when the call frequency decreases, the input of computing resources will be reduced to avoid waste of resources.

[0082] In order to achieve load balancing, in one embodiment of the present invention, the step 103 in the above embodiment of determining the load of at least one node according to the usage of system resources includes:

[0083] Determine whether there is a target node among the at least one node whose load condition is lower than a first preset threshold, wherein the first preset threshold is used to represent a minimum value of the load condition of the at least one node;

[0084] When the load condition of the target node among the at least one node is lower than the first preset threshold, the task allocation amount of the target node is increased.

[0085] In an embodiment of the present invention, a deep learning algorithm for load balancing can be used to implement intelligent scheduling of resources. It can monitor the load conditions of each node. When it is detected that the load condition of a certain node, namely the target node, is lower than a preset first preset threshold, it automatically allocates tasks to the target node with a lower load, thereby achieving load balancing, improving the overall performance of the system, and ultimately realizing dynamic allocation of resources.

[0086] In order to realize intuitive visualization of resources, in one embodiment of the present invention, in the above embodiment, after performing resource management based on the resource demand conditions of the static resources and the dynamic resources and the load conditions of the at least one node by using the intelligent AI management platform in step 104, the method further includes:

[0087] Classifying the computing resources and generating target computing resources, wherein the target computing resources include at least one of a central processing unit (CPU), a cloud disk, and a network;

[0088] Based on the target computing resource, determine the usage and demand of at least one of the central processing unit CPU, the cloud disk and the network;

[0089] Based on the usage and demand conditions of at least one of the central processing unit CPU, cloud disk and network, a dynamic resource allocation visualization VR system is constructed, wherein the dynamic resource allocation visualization VR system is used to intuitively visualize resource allocation.

[0090] In an embodiment of the present invention, building a cloud-edge collaborative dynamic resource allocation visualization VR system requires not only the ability to accurately allocate resources dynamically according to resource requirements, but also the intuitive visualization of resource allocation, so that users can understand the usage and allocation status of resources in real time. In terms of resource refinement, the system will classify and define various computing resources in detail, including central processing unit CPU, memory, storage and network bandwidth. Through precise refinement, the system can more accurately grasp the usage and demand of each resource, so as to carry out more refined management. Dynamic resource allocation is the core function of this system. The system will dynamically adjust the allocation of resources according to real-time resource requirements and usage. When an application or task needs more resources, the system will automatically schedule resources from other idle or low-load nodes; when the resource demand decreases, the system will release the excess resources back to the resource pool for use by other applications or tasks. Through cloud-edge collaboration, the system can make full use of the computing resources of the cloud and edge to achieve resource complementarity and optimization. The cloud can provide powerful computing power and storage resources, while the edge can provide lower latency and better real-time performance. By working together, the system can reduce operating costs and improve the overall performance of the system while meeting application requirements. The visualization VR system is an intuitive display of this technology. Through virtual reality technology, users can intuitively see the distribution, usage status and allocation of various resources in a three-dimensional space. This can not only improve users' understanding of resource usage, but also help users better manage and optimize resources. Based on resource-specific dynamic resource allocation, building a cloud-edge collaborative dynamic resource allocation visualization VR system is an innovative and practical technical project. It will help improve resource utilization efficiency and management level, and create greater value for enterprises and users.

[0091] In order to alleviate the fatigue state of the hardware system, in one embodiment of the present invention, in the above embodiment, after the resource demand situation of the static resources and the dynamic resources and the load situation of the at least one node are used to manage resources by the intelligent AI management platform in step 104, the method further includes:

[0092] Using sensors deployed on the hardware system to collect operating data of the hardware system in real time, wherein the operating data includes at least one of historical data, real-time operating data, and external environmental factor data of the hardware system;

[0093] Based on the operating data, determining whether the hardware system is in a fatigue state;

[0094] When it is determined that the hardware system is in the fatigue state, operating parameters of the hardware system are adjusted, wherein the operating parameters are used to alleviate the fatigue state of the hardware system.

[0095] In the embodiment of the present invention, the model fusion can be used to dynamically alleviate the fatigue of the hardware system and the implementation scheme of the resource recovery mechanism, which integrates advanced algorithms, cloud computing and edge computing technologies. The scheme aims to monitor the operating status of the hardware system in real time through intelligent means, predict and alleviate its fatigue level, and realize the effective recovery and reuse of resources, thereby improving the stability of the system and the efficiency of resource utilization.

[0096] Specifically, in terms of dynamic model fusion, multiple algorithms and models can be used to work together to achieve accurate perception and prediction of the hardware system status. This includes but is not limited to machine learning models, deep learning models, statistical analysis models, and system simulation models. These models will conduct a comprehensive analysis of the hardware system's historical data, real-time operation data, and external environmental factors to derive the system's current fatigue level and possible future development trends. Once the model predicts that the hardware system is showing signs of fatigue, we will immediately start a dynamic adjustment mechanism. This mechanism will automatically adjust the system's operating parameters, optimize task allocation, or take other necessary intervention measures based on the model's prediction results to alleviate the system's fatigue state. These adjustment measures are designed to balance the system's workload, reduce energy consumption and heat generation, and thus extend the service life of the hardware. At the same time, a cloud-edge collaborative resource recovery mechanism will also be built. At the cloud computing level, a unified resource management platform will be established to centrally store, manage, and schedule recycled resources. Through the powerful data processing capabilities and elastic scalability of cloud computing, the rapid integration and reuse of recycled resources can be achieved. At the edge computing level, the real-time perception and response capabilities of edge devices are used to achieve real-time monitoring and rapid response to the use of hardware system resources. Once idle or inefficient resources are found, the edge device will immediately trigger the resource recovery mechanism and recycle these resources to the resource management platform for unified processing. Through cloud-edge collaboration, accurate matching and efficient use of resources can be achieved. The centralized management and scheduling capabilities of cloud computing can ensure that resources are reasonably allocated and utilized, while the real-time perception and response capabilities of edge computing can ensure that resources are promptly recovered and reused. In summary, the implementation scheme of the dynamic alleviation of hardware system fatigue and resource recovery mechanism based on model fusion is a comprehensive solution that combines advancement and practicality. It can not only improve the stability and reliability of the hardware system and extend its service life, but also improve the utilization efficiency of resources, reduce operating costs, and create greater value for enterprises and users.

[0097] like Figure 2 As shown, in order to more clearly illustrate the technical solution and advantages of the present invention, a resource management method based on cloud-edge collaboration is provided in detail for an embodiment of the present invention, which may specifically include the following steps:

[0098] Step 201: Building an intelligent AI management platform based on server performance data, network traffic data, and storage usage data, wherein the intelligent AI management platform is used to manage computing, network, and storage resources;

[0099] Specifically, the core of the intelligent AI management platform is the resource management algorithm. This algorithm adopts a physical structure resource pooling model to build a bionic human model. The computing power resource pool is simulated as the "brain", the network is simulated as the "blood", and the storage is simulated as the "limbs". The "fatigue" level of each resource pool is collected in real time. The "cerebellum" of the bionic human management system is responsible for regulating the balance of the storage "limbs", and the brain stem needs to maintain the operation of the network "blood" and replenish it in time to where it is needed, so that the resource link is reachable. According to the storage performance and the IO required by the program, high, low and medium performance storage and network traffic and other parameter configurations are allocated.

[0100] For example, the intelligent AI management platform needs to obtain multi-dimensional data including server performance data, network traffic data, storage usage, etc. These data can be collected in real time or regularly through log collection tools, API interfaces, etc. By analyzing and predicting these data, the intelligent AI management platform can automatically adjust resource allocation strategies, optimize resource utilization, reduce costs, and improve the performance and reliability of the overall system. This intelligent management method makes resource allocation more accurate and efficient, thereby meeting the needs of business development.

[0101] Step 202: Determine static resource requirements based on periodic changes, seasonal changes, and the impact of emergencies in static resource usage, where static resources refer to fixed files stored on the server;

[0102] Step 203: Determine whether the frequency of interface calls of dynamic resources increases, wherein dynamic resources are contents dynamically generated on the server side;

[0103] Step 204: when the frequency of interface calls of dynamic resources increases, increase the investment of computing resources;

[0104] Step 205: When the frequency of interface calls of dynamic resources decreases, reducing the input of computing resources;

[0105] Specifically, a deep learning algorithm based on load balancing can be used, which can identify the periodicity, burstiness and correlation of static resource usage, thereby predicting static resource demand. In this way, resource allocation can be adjusted to ensure increased resource supply under high load and effective resource recovery under low load, so as to maximize resource utilization and stable system operation.

[0106] Step 206: Determine whether there is a target node in at least one node whose load is lower than a first preset threshold value according to the usage of system resources, wherein the usage of system resources includes at least one of a CPU usage rate, a memory occupancy rate, and a network bandwidth, and the first preset threshold value is used to represent a minimum value of the load of at least one node;

[0107] Step 207: when the load of the target node in at least one node is lower than the first preset threshold, increase the task allocation amount of the target node;

[0108] Step 208: Based on the resource demand of static resources and dynamic resources and the load of at least one node, use the intelligent AI management platform to manage resources;

[0109] Step 209: Classify the computing resources and generate target computing resources, wherein the target computing resources include at least one of a central processing unit CPU, a cloud disk, and a network;

[0110] Step 210: Determine the usage and demand of at least one of a central processing unit (CPU), a cloud disk, and a network based on the target computing resource;

[0111] Step 211: constructing a dynamic resource allocation visualization VR system based on the usage and demand of at least one of a central processing unit (CPU), a cloud disk, and a network, wherein the dynamic resource allocation visualization VR system is used to visualize resource allocation intuitively;

[0112] Step 212: using sensors deployed on the hardware system to collect operating data of the hardware system in real time, wherein the operating data includes at least one of historical data of the hardware system, real-time operating data, and external environmental factor data;

[0113] Step 213: Determine whether the hardware system is in a fatigue state based on the operation data;

[0114] Specifically, when predicting the degree of hardware system fatigue, the historical data of the hardware system (such as past performance, fault records, etc.), real-time operation data (such as current CPU load, temperature, voltage fluctuations, etc.) and external environmental factors (such as ambient temperature, humidity, dust accumulation, etc.) are comprehensively considered. Through in-depth analysis of these data, the model can identify patterns of system performance degradation, abnormal fluctuations, or reaching specific thresholds. These patterns are usually associated with hardware fatigue or the imminent fatigue state. When the model detects that the system performance indicators are continuously below the normal threshold, the fluctuations increase, or abnormal patterns appear, it will be judged that the system is in a fatigue state or is about to enter a fatigue state. On the contrary, if the system performance indicators are stable and remain within the normal range, and do not show abnormal patterns, the model will judge that the system is not currently in a fatigue state. This comprehensive analysis makes the prediction more accurate and helps to take timely measures to alleviate the fatigue of the hardware system;

[0115] Step 214: When it is determined that the hardware system is in a fatigue state, adjust operating parameters of the hardware system, wherein the operating parameters are used to alleviate the fatigue state of the hardware system.

[0116] For example, in order to alleviate the fatigue state of the hardware system, the automatic adjustment plan may involve adjusting key operating parameters such as CPU frequency, memory allocation strategy, disk I / O priority, etc. to reduce system load and heat generation. At the same time, optimizing task allocation is also an important part. Through intelligent scheduling algorithms, high-load tasks are distributed to different processors or time periods to avoid excessive use of a single component. In addition, intervention measures such as starting hot-swappable backup hardware, adjusting cooling strategies, or starting resource recovery mechanisms may be taken to further balance the system load, extend hardware life, and ensure overall system performance and stability.

[0117] like Figure 3 As shown, an embodiment of the present invention provides a resource management system based on cloud-edge collaboration, the system comprising:

[0118] A platform construction module 301 is used to construct an intelligent AI management platform based on server performance data, network traffic data, and storage usage data, wherein the intelligent AI management platform is used to perform resource management on computing, network, and storage;

[0119] The resource demand determination module 302 is used to determine the resource demand of static resources and dynamic resources, wherein the static resources refer to fixed files stored on the server, and the dynamic resources refer to content dynamically generated on the server side;

[0120] The load determination module 303 is used to determine the load of at least one node according to the usage of system resources, wherein the usage of system resources includes at least one of: CPU usage, memory occupancy and network bandwidth;

[0121] The resource management module 304 is used to perform resource management using the intelligent AI management platform based on the resource demand conditions of the static resources and the dynamic resources and the load conditions of the at least one node.

[0122] based on Figure 3 In the resource management system based on cloud-edge collaboration shown in the embodiment, the resource demand determination module 302 in the above embodiment is also used to execute:

[0123] Determining the static resource demand based on the periodic changes, seasonal changes and impact of emergencies in the static resource usage;

[0124] Determining whether the interface call frequency of the dynamic resource increases;

[0125] When the frequency of interface calls of the dynamic resource increases, increasing the investment of computing resources;

[0126] When the frequency of interface calls of the dynamic resource decreases, the investment in the computing resources is reduced.

[0127] based on Figure 3 In the resource management system based on cloud-edge collaboration shown in the figure, the load determination module 303 in the above embodiment is also used to execute:

[0128] Determine whether there is a target node among the at least one node whose load condition is lower than a first preset threshold, wherein the first preset threshold is used to represent a minimum value of the load condition of the at least one node;

[0129] When the load condition of the target node among the at least one node is lower than the first preset threshold, the task allocation amount of the target node is increased.

[0130] based on Figure 3 A resource management system based on cloud-edge collaboration is shown in Figure 4 As shown, in one embodiment of the present invention, after the resource management module 304, it further includes: a resource visualization module 305,

[0131] The resource visualization module 305 is used to perform:

[0132] Classifying the computing resources and generating target computing resources, wherein the target computing resources include at least one of a central processing unit (CPU), a cloud disk, and a network;

[0133] Based on the target computing resource, determine the usage and demand of at least one of the central processing unit CPU, the cloud disk and the network;

[0134] Based on the usage and demand of at least one of the central processing unit CPU, the cloud disk and the network, a dynamic resource allocation visualization VR system is constructed, wherein the dynamic resource allocation visualization VR system is used to intuitively visualize resource allocation;

[0135] based on Figure 4 A resource management system based on cloud-edge collaboration is shown in Figure 5 As shown, in one embodiment of the present invention, after the resource visualization module 305, a fatigue processing module 306 is further included;

[0136] The fatigue processing module 306 is used to execute:

[0137] Using sensors deployed on the hardware system to collect operating data of the hardware system in real time, wherein the operating data includes at least one of historical data, real-time operating data, and external environmental factor data of the hardware system;

[0138] Based on the operating data, determining whether the hardware system is in a fatigue state;

[0139] When it is determined that the hardware system is in the fatigue state, operating parameters of the hardware system are adjusted, wherein the operating parameters are used to alleviate the fatigue state of the hardware system.

[0140] It is to be understood that the structure illustrated in the embodiment of the present invention does not constitute a specific limitation on a report merging device. In other embodiments of the present invention, a report merging device may include more or fewer components than shown in the figure, or combine some components, or split some components, or arrange the components differently. The components shown in the figure may be implemented in hardware, software, or a combination of software and hardware.

[0141] The information interaction, execution process and other contents between the units in the above-mentioned device are based on the same concept as the embodiment of the method of the present invention. For specific contents, please refer to the description in the embodiment of the method of the present invention, and no further description is given here.

[0142] An embodiment of the present invention further provides a resource management system based on cloud-edge collaboration, comprising: at least one memory and at least one processor;

[0143] at least one memory for storing a machine-readable program;

[0144] At least one processor is used to call a machine-readable program to execute a resource management method based on cloud-edge collaboration in any embodiment of the present invention.

[0145] An embodiment of the present invention also provides a computer-readable medium, on which computer instructions are stored. When the computer instructions are executed by a processor, the processor executes a resource management method based on cloud-edge collaboration in any embodiment of the present invention.

[0146] Specifically, a system or device equipped with a storage medium can be provided, on which software program code that implements the functions of any of the above-mentioned embodiments is stored, and a computer (or CPU or MPU) of the system or device can be enabled to read and execute the program code stored in the storage medium.

[0147] In this case, the program code itself read from the storage medium can realize the function of any one of the above-mentioned embodiments, and thus the program code and the storage medium storing the program code constitute a part of the present invention.

[0148] The storage medium embodiments for providing the program code include a floppy disk, a hard disk, a magneto-optical disk, an optical disk (such as CD-ROM, CD-R, CD-RW, DVD-ROM, DVD-RAM, DVD-RW, DVD+RW), a magnetic tape, a non-volatile memory card, and a ROM. Alternatively, the program code can be downloaded from a server computer by a communication network.

[0149] In addition, it should be clear that the functions of any of the above embodiments can be implemented not only by executing the program code read by the computer, but also by enabling an operating system operating on the computer to complete part or all of the actual operations based on instructions from the program code.

[0150] In addition, it can be understood that the program code read from the storage medium is written to a memory provided in an expansion board inserted into the computer or written to a memory provided in an expansion unit connected to the computer, and then based on the instructions of the program code, a CPU installed on the expansion board or the expansion unit is enabled to perform part or all of the actual operations, thereby realizing the functions of any of the above-mentioned embodiments.

[0151] Each embodiment of the present invention has at least the following beneficial effects:

[0152] 1. In an embodiment of the present invention, a resource management method based on cloud-edge collaboration is provided. In the edge computing scenario, the contradiction between the task's thirst for resources and the scarcity of edge device resources is prominent. The computing and storage capabilities of edge nodes are relatively weak compared to those of the cloud. In addition, edge nodes also contain a variety of heterogeneous nodes and are geographically dispersed, and the bandwidth resources of edge networks are relatively scarce. Therefore, the reasonable allocation of limited resources is the key difficulty in achieving low power consumption and low latency in the cloud-edge collaborative architecture. In order to solve the above difficulties, an intelligent AI management platform can be first constructed to better allocate computing, storage and network resources. Then, by monitoring the call frequency of the data interface of static resources and dynamic resources, the resource requirements of static resources and dynamic resources can be accurately estimated, and then the dynamic allocation of computing resources can be realized, which can help enterprises better manage and utilize computing resources, improve the stability and reliability of the system, and reduce operating costs, creating greater value for enterprises. At the same time, intelligent scheduling of resources can be realized based on load balancing. It can automatically allocate tasks to nodes with lower loads according to the load conditions of each node, thereby achieving load balancing and improving the overall performance of the system. Through the above-mentioned dynamic resource allocation process, resource allocation can be made more reasonable;

[0153] 2. In the embodiment of the present invention, the resource requirements of static resources and dynamic resources can be accurately estimated based on the deep learning algorithm of load balancing. This algorithm can analyze the usage of system resources in real time, and predict the resource requirements in the future through deep learning technology, so as to achieve efficient management and allocation of resources. First, for static resources, the deep learning algorithm will use its powerful data processing and learning capabilities to analyze a large amount of historical data to find out the rules of static resource usage. This includes modeling the periodic changes, seasonal changes and the impact of emergencies in resource usage. Through these models, the deep learning algorithm can predict the demand for static resources in the future, so as to prepare and allocate static resources in advance. For dynamic resources, the deep learning algorithm pays more attention to real-time and dynamic. It will monitor the call frequency of the data interface in real time and dynamically adjust the allocation of computing resources according to the changes in the call frequency. When the call frequency increases, the algorithm will automatically increase the input of computing resources to meet the system processing requirements; when the call frequency decreases, it will reduce the input of computing resources to avoid waste of resources;

[0154] 3. In the embodiment of the present invention, the construction of a cloud-edge collaborative dynamic resource allocation visualization VR system requires not only the ability to accurately allocate resources dynamically according to resource requirements, but also the intuitive visualization of resource allocation, so that users can understand the usage and allocation status of resources in real time. In terms of resource refinement, the system will classify and define various computing resources in detail, including central processing unit CPU, memory, storage and network bandwidth. Through precise refinement, the system can more accurately grasp the usage and demand of each resource, so as to carry out more refined management. Dynamic resource allocation is the core function of this system. The system will dynamically adjust the allocation of resources according to real-time resource requirements and usage. When an application or task needs more resources, the system will automatically schedule resources from other idle or low-load nodes; when the resource demand decreases, the system will release the excess resources back to the resource pool for use by other applications or tasks. Through cloud-edge collaboration, the system can make full use of the computing resources of the cloud and edge to achieve resource complementarity and optimization. The cloud can provide powerful computing power and storage resources, while the edge can provide lower latency and better real-time performance. By working together, the system can reduce operating costs and improve the overall performance of the system while meeting application requirements. The visualization VR system is an intuitive display of this technology. Through virtual reality technology, users can intuitively see the distribution, usage status and allocation of various resources in a three-dimensional space. This can not only improve users' understanding of resource usage, but also help users better manage and optimize resources. Based on resource-specific dynamic resource allocation, building a cloud-edge collaborative dynamic resource allocation visualization VR system is an innovative and practical technical project. It will help improve resource utilization efficiency and management level, and create greater value for enterprises and users.

[0155] It should be noted that not all steps and modules in the above-mentioned processes and system structure diagrams are necessary, and some steps or modules can be ignored according to actual needs. The execution order of each step is not fixed and can be adjusted as needed. The system structure described in the above-mentioned embodiments can be a physical structure or a logical structure, that is, some modules may be implemented by the same physical entity, or some modules may be implemented by multiple physical entities, or some components in multiple independent devices may be implemented together.

[0156] In the above embodiments, the hardware unit can be realized by mechanical means or electrical means. For example, a hardware unit can include permanent dedicated circuits or logic (such as special processors, FPGA or ASIC) to complete the corresponding operation. The hardware unit can also include programmable logic or circuits (such as general-purpose processors or other programmable processors), which can be temporarily set by software to complete the corresponding operation. Concrete implementation (mechanical means or dedicated permanent circuits or temporarily set circuits) can be determined based on cost and time considerations.

[0157] The above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions, improvements, etc. made within the spirit and principles of the present invention should be included in the scope of protection of the present invention.

Claims

1. A resource management method based on cloud-edge collaboration, characterized in that: The method includes: Building an intelligent AI management platform based on server performance data, network traffic data, and storage usage data, wherein the intelligent AI management platform is used to manage computing, network, and storage resources; Determine resource requirements for static resources and dynamic resources, wherein the static resources refer to fixed files stored on the server and the dynamic resources refer to content dynamically generated on the server side; Determine the load of at least one node according to the usage of system resources, wherein the usage of system resources includes at least one of: CPU usage, memory occupancy, and network bandwidth; Based on the resource demand conditions of the static resources and the dynamic resources and the load conditions of at least one node, the intelligent AI management platform is used to perform resource management.

2. The method according to claim 1, characterized in that The determining of resource requirements of static resources and dynamic resources includes: Determining the static resource demand based on the periodic changes, seasonal changes and impact of emergencies in the static resource usage; Determining whether the interface call frequency of the dynamic resource increases; When the frequency of interface calls of the dynamic resource increases, increasing the investment of computing resources; When the frequency of interface calls of the dynamic resource decreases, the investment in the computing resources is reduced.

3. The method according to claim 1, characterized in that The step of determining the load of at least one node according to the usage of system resources includes: Determine whether there is a target node among the at least one node whose load condition is lower than a first preset threshold, wherein the first preset threshold is used to represent a minimum value of the load condition of the at least one node; When the load condition of the target node among the at least one node is lower than the first preset threshold, the task allocation amount of the target node is increased.

4. The method according to any one of claims 1 to 3, characterized in that: After the resource demand conditions of the static resources and the dynamic resources and the load conditions of the at least one node are used to manage resources using the intelligent AI management platform, the method further includes: Classifying the computing resources and generating target computing resources, wherein the target computing resources include at least one of a central processing unit (CPU), a cloud disk, and a network; Based on the target computing resource, determine the usage and demand of at least one of the central processing unit CPU, the cloud disk and the network; Based on the usage and demand of at least one of the central processing unit CPU, the cloud disk and the network, a dynamic resource allocation visualization VR system is constructed, wherein the dynamic resource allocation visualization VR system is used to intuitively visualize resource allocation; and / or, Using sensors deployed on the hardware system to collect operating data of the hardware system in real time, wherein the operating data includes at least one of historical data, real-time operating data, and external environmental factor data of the hardware system; Based on the operating data, determining whether the hardware system is in a fatigue state; When it is determined that the hardware system is in the fatigue state, operating parameters of the hardware system are adjusted, wherein the operating parameters are used to alleviate the fatigue state of the hardware system.

5. A resource management system based on cloud-edge collaboration, characterized in that: The system includes: A platform building module, used to build an intelligent AI management platform based on server performance data, network traffic data, and storage usage data, wherein the intelligent AI management platform is used to perform resource management on computing, network, and storage; A resource demand determination module, used to determine the resource demand of static resources and dynamic resources, wherein the static resources refer to fixed files stored on the server, and the dynamic resources refer to content dynamically generated on the server side; A load determination module, configured to determine the load of at least one node according to the usage of system resources, wherein the usage of system resources includes at least one of: CPU usage, memory occupancy, and network bandwidth; A resource management module is used to perform resource management using the intelligent AI management platform based on the resource demand conditions of the static resources and the dynamic resources and the load conditions of at least one node.

6. The system according to claim 5, characterized in that The resource requirement determination module is further used to execute: Determining the static resource demand based on the periodic changes, seasonal changes and impact of emergencies in the static resource usage; Determining whether the interface call frequency of the dynamic resource increases; When the frequency of interface calls of the dynamic resource increases, increasing the investment of computing resources; When the frequency of interface calls of the dynamic resource decreases, the investment in the computing resources is reduced.

7. The system according to claim 5, characterized in that The load determination module is further configured to execute: Determine whether there is a target node among the at least one node whose load condition is lower than a first preset threshold, wherein the first preset threshold is used to represent a minimum value of the load condition of the at least one node; When the load condition of the target node among the at least one node is lower than the first preset threshold, the task allocation amount of the target node is increased.

8. The system according to any one of claims 5 to 7, characterized in that: After the resource management, it further includes: a resource visualization module, The resource visualization module is used to perform: Classifying the computing resources and generating target computing resources, wherein the target computing resources include at least one of a central processing unit (CPU), a cloud disk, and a network; Based on the target computing resource, determine the usage and demand of at least one of the central processing unit CPU, the cloud disk and the network; Based on the usage and demand of at least one of the central processing unit CPU, the cloud disk and the network, a dynamic resource allocation visualization VR system is constructed, wherein the dynamic resource allocation visualization VR system is used to intuitively visualize resource allocation; and / or, Further comprising: a fatigue processing module; The fatigue processing module is used to execute: Using sensors deployed on the hardware system to collect operating data of the hardware system in real time, wherein the operating data includes at least one of historical data, real-time operating data, and external environmental factor data of the hardware system; Based on the operating data, determining whether the hardware system is in a fatigue state; When it is determined that the hardware system is in the fatigue state, operating parameters of the hardware system are adjusted, wherein the operating parameters are used to alleviate the fatigue state of the hardware system.

9. A resource management system based on cloud-edge collaboration, characterized in that: include: at least one memory and at least one processor; The at least one memory is used to store a machine-readable program; The at least one processor is configured to call the machine-readable program to execute the method according to any one of claims 1 to 4.

10. A computer readable medium, characterized in that The computer readable medium stores computer instructions, which, when executed by a processor, cause the processor to perform any one of the methods of claims 1 to 4.

Images