A resource allocation management method and computer equipment for cloud servers

Abstract

The present invention relates to the field of electronic digital data processing technology, specifically to a resource allocation management method and computer equipment for a cloud server, the method comprising: collecting resource demand data of each user group, including basic resource usage data and business data; comprehensively analyzing to obtain a predicted resource demand evaluation index and a resource allocation frequency reference index for each user group, and pre-allocating resources, monitoring the load data of each user group, comprehensively analyzing to obtain a resource utilization evaluation index for each user group; and redistributing resources to each user group resource pool according to the resource allocation frequency reference index and the resource utilization evaluation index for each user group. The present invention provides a resource allocation management method and computer equipment for a cloud server, thereby ensuring that the resources of the cloud server can be reasonably allocated, helping to achieve load balancing, and ensuring the stability of the service. It also helps to improve service quality, reduce delays, and provide a smoother user experience.

Description

Technical Field

[0001] The present invention relates to the technical field of electronic digital data processing, and in particular to a resource allocation management method and computer equipment of a cloud server. Background Art

[0002] With the popularization of the Internet and the acceleration of digital transformation, enterprises and society have generated massive amounts of data. From social media, e-commerce to IoT devices, the explosive growth of data has put unprecedented demands on data processing and storage capabilities. Traditional information technology infrastructure is difficult to adapt to such volatile demand. Cloud computing provides an elastic solution that can dynamically adjust resources according to actual needs, avoiding the waste and insufficiency caused by fixed resource investment.

[0003] For example, the invention patent with announcement number: CN110532087B is a resource allocation method and resource manager, wherein the method includes: obtaining a resource scheduling request; responding to the scheduling request, determining the resource quota of each user group according to a preset strategy, wherein the preset strategy is: equally distributing the total cluster resources among the user groups, and distributing the resources exceeding the demand of the user group after the equal distribution equally among the user groups with insufficient demand; and allocating resources to each user group according to the determined resource quota of each user group. The technical solution provided by the embodiment of the application solves the technical problem that resources cannot be fairly distributed among different user groups in the existing method of traversing the queue and allocating the queue requests in the queue in sequence, and achieves the technical effect of simply and efficiently realizing the fair distribution of resources among different user groups.

[0004] For example, the invention patent with announcement number: CN107045456B is a resource allocation method and resource manager for improving resource utilization and / or for improving the execution efficiency of user jobs. The method includes: receiving a job submitted by a client device, and decomposing the job into multiple tasks, wherein each of the multiple tasks is configured with a corresponding resource requirement; estimating the running time of each task; according to the resource requirement and running time corresponding to each task, combined with a preset scheduling strategy, determining the first allocation configuration of the multiple tasks, the first allocation configuration is used to indicate the distribution of the multiple tasks on the executable computing nodes in the multiple computing nodes, the scheduling strategy includes at least one of a resource utilization priority strategy and an efficiency priority strategy; allocating the multiple tasks to the executable computing nodes of the multiple tasks according to the first allocation configuration.

[0005] However, in the process of implementing the technical solution of the invention in the embodiments of the present application, the present application discovered that the above technology has at least the following technical problems: the current resource allocation management method includes automatically allocating resources equally among user groups in a cloud server, and then performing secondary allocation based on real-time demand, but this often sacrifices efficiency due to a focus on fairness and has low flexibility, which may result in waste or shortage of resources. Summary of the invention

[0006] In view of the deficiencies in the prior art, the present invention provides a resource allocation management method and computer device for a cloud server, which can effectively solve the problems involved in the above-mentioned background technology.

[0007] To achieve the above objectives, the present invention is implemented through the following technical solutions: The first aspect of the present invention provides a resource allocation management method for a cloud server, including: collecting resource demand data of each user group, wherein the resource demand data includes basic resource usage data and business data.

[0008] Based on the basic resource usage data, a comprehensive analysis is performed to obtain the predicted resource demand assessment index of each user group, and resources are pre-allocated based on the predicted resource demand assessment index of each user group. The pre-allocated resources for each user group are marked as the resource pool of each user group.

[0009] Based on the business data, a comprehensive analysis is performed to obtain a reference index for resource allocation frequency of each user group, and the load data of each user group is monitored and a comprehensive analysis is performed to obtain an evaluation index for resource utilization of each user group.

[0010] The resource pools of each user group are reallocated according to the resource allocation frequency reference index and resource utilization evaluation index of each user group.

[0011] As a further method, the comprehensive analysis obtains the predicted resource demand evaluation index of each user group, and the specific analysis process is: deploying a number of time monitoring points, dividing the time monitoring points into historical time monitoring points and current time monitoring points according to a preset time scale, collecting basic resource usage data of each user group at each historical time monitoring point and each current time monitoring point, extracting the critical CPU usage rate, critical memory usage and critical disk space occupancy from the cloud server database, and obtaining the historical basic resource demand evaluation index and the current basic resource demand evaluation index of each user group after processing;

[0012] Based on the historical basic resource demand assessment index and the current basic resource demand assessment index of each user group, a comprehensive analysis is performed to obtain the predicted resource demand assessment index of each user group.

[0013] As a further method, resource pre-allocation is performed based on the predicted resource demand level assessment index of each user group. The specific allocation process is: comparing the predicted resource demand level assessment index of each user group with the cloud server resource demand corresponding to each predicted resource demand level assessment index interval stored in the cloud server database, obtaining the cloud server resource demand of each user group, and allocating resources to each user group based on the cloud server resource demand of each user group.

[0014] As a further method, the comprehensive analysis obtains the resource allocation frequency reference index of each user group, and the specific analysis process is: collecting the business data of each user group at each historical time monitoring point and each current time monitoring point, extracting the critical transaction volume and critical access volume from the cloud server database, and obtaining the historical business demand evaluation index and current business demand evaluation index of each user group after processing;

[0015] According to the historical business demand assessment index and the current business demand assessment index of each user group, a comprehensive analysis is performed to obtain the resource allocation frequency reference index of each user group.

[0016] As a further method, the comprehensive analysis obtains the resource utilization evaluation index of each user group. The specific analysis process is: deploying several time recording points, monitoring the load data of each user group in real time, wherein the load data includes the number of concurrent requests, response time and throughput of each user group at each time recording point, extracting the critical concurrent request number, critical response time and critical throughput from the cloud server database, and comprehensively analyzing to obtain the resource utilization evaluation index of each user group.

[0017] As a further method, the resource pool of each user group is reallocated according to the resource allocation frequency reference index and resource utilization evaluation index of each user group. The specific allocation process is: comparing the resource allocation frequency reference index of each user group with the resource allocation frequency reference index threshold preset interval stored in the cloud server database. If the resource allocation frequency reference index of a user group is higher than the upper limit of the resource allocation frequency reference index preset interval or lower than the lower limit of the resource allocation frequency reference index preset interval, the user group is marked as a demand reallocation user group;

[0018] Obtain resource utilization evaluation indexes of each demand redistribution user group, sort them in ascending order, and record the resource utilization evaluation index in the binary sequence as the reference standard resource utilization evaluation index;

[0019] Subtract the resource utilization degree evaluation index of each demand redistribution user group from the reference standard resource utilization degree evaluation index, and mark the absolute value of the difference result as the resource redistribution reference index of each demand redistribution user group. Then, perform resource redistribution for each demand redistribution user group according to the resource redistribution reference index of each demand redistribution user group.

[0020] As a further method, the resource utilization degree evaluation index of each user group is quantitative evaluation data obtained through comprehensive analysis of the number of concurrent requests, response time, and throughput of each user group at each time recording point, and is used to quantitatively evaluate the resource utilization degree of each user group, providing a basis for reasonably allocating cloud server resources.

[0021] As a further method, the numerical expression of the resource allocation frequency reference index of each user group is:

[0022] W = tanh(τQn * 81 + TKn * ε2);

[0023] In the formula, W n represents the resource allocation frequency reference index of the nth user group, τQ n represents the historical business demand evaluation index of the nth user group, Ms n represents the trading volume of the nth user group at the s-th historical time monitoring point, Ns n represents the number of visits of the nth user group at the s-th historical time monitoring point, s represents the number of each historical time monitoring point, s = 1, 2, 3,..., h, and h represents the total number of historical time monitoring points, τK n represents the historical business demand evaluation index of the nth user group, Min represents the trading volume of the nth user group at the i-th current time monitoring point, Ni n represents the number of visits of the nth user group at the i-th current time monitoring point, i represents the number of each current time monitoring point, i = 1, 2, 3,..., t, and t represents the total number of current time monitoring points, represents the critical trading volume, represents the critical number of visits, represents the historical business demand evaluation impact factor corresponding to the preset trading volume, represents the historical business demand evaluation impact factor corresponding to the preset number of visits, represents the current business demand evaluation impact factor corresponding to the preset trading volume, represents the current business demand evaluation impact factor corresponding to the preset number of visits, ε1 represents the resource allocation frequency impact factor corresponding to the preset historical business demand evaluation index, and ε2 represents the resource allocation frequency impact factor corresponding to the preset current business demand evaluation index.

[0024] As a further method, the numerical expression of the resource utilization evaluation index of each user group is:

[0025]

[0026] Where, L n represents the resource utilization evaluation index of the nth user group, q represents the number of each time recording point, q=1,2,3,...,g, g represents the total number of time recording points, e represents a natural constant, α nq represents the number of concurrent requests of the nth user group at the qth time point, β nq represents the throughput of the nth user group at the qth time recording point, Indicates the critical number of concurrent requests. represents the critical throughput, θ1 represents the resource utilization assessment impact factor corresponding to the set number of concurrent requests, and θ2 represents the resource utilization assessment impact factor corresponding to the set throughput.

[0027] The second aspect of the present invention provides a resource allocation management computer device for a cloud server, characterized in that it includes: a processor and a memory and a network interface connected to the processor; the network interface is connected to a non-volatile memory in the server; the processor calls a computer program from the non-volatile memory through the network interface during operation, and runs the computer program through the memory to execute any one of the methods described in the above inventions.

[0028] Compared with the prior art, the embodiments of the present invention have at least the following advantages or beneficial effects:

[0029] (1) The present invention provides a cloud server resource allocation management method and computer equipment to ensure that cloud server resources can be reasonably allocated, help achieve load balancing, prevent some servers from being overloaded, ensure service stability and response time, and help establish an effective disaster recovery plan to ensure that sufficient backup resources can be quickly switched in the event of a failure to maintain business continuity. Reasonable allocation of cloud server resources can improve service quality, reduce delays, and provide a smoother user experience.

[0030] (2) The present invention helps to better plan future resource needs by predicting resource needs and establishing resource pools, so that the system can make adjustments in advance to cope with business growth or seasonal changes, achieve resource isolation and protection, and ensure resource stability and security. By pre-allocating resources, the resource pool can ensure that each user group has sufficient resources to run its applications and services, avoiding performance degradation caused by resource competition.

[0031] (3) The present invention can guide the implementation of dynamic adjustment strategies by analyzing resource allocation frequency. This immediate response mechanism can improve resource utilization efficiency while maintaining good service quality and user experience. Predicting resource demand trends can ensure that there are sufficient resources to support peak demand periods. By analyzing resource demand frequency, additional resources can be allocated in advance to prevent service interruptions or performance degradation, ensuring the smooth operation of key business processes.

[0032] (4) By analyzing the utilization of resources, the present invention can avoid over-allocation of resources, reduce unnecessary costs, and help support dynamic resource allocation strategies, ensuring that resource usage matches actual needs and further optimizing costs. At the same time, analyzing the utilization of resources can help identify performance bottlenecks and improve overall system performance. BRIEF DESCRIPTION OF THE DRAWINGS

[0033] The present invention is further described using the accompanying drawings, but the embodiments in the accompanying drawings do not constitute any limitation to the present invention. A person skilled in the art can obtain other drawings based on the following drawings without creative work.

[0034] Figure 1 It is a schematic diagram of the method flow of the present invention.

[0035] Figure 2 It is a schematic diagram of the functional relationship between the historical basic resource demand assessment index and the predicted resource demand degree assessment index involved in an embodiment of the present invention. DETAILED DESCRIPTION

[0036] 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 only part of the embodiments of the present invention, rather than all the embodiments. Based on the embodiments of 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.

[0037] Reference Figure 1 As shown, the first aspect of the present invention provides a resource allocation management method for a cloud server, comprising: collecting resource demand data of each user group, wherein the resource demand data comprises basic resource usage data and business data.

[0038] Based on the basic resource usage data, a comprehensive analysis is performed to obtain the predicted resource demand assessment index of each user group, and resources are pre-allocated based on the predicted resource demand assessment index of each user group. The pre-allocated resources for each user group are marked as the resource pool of each user group.

[0039] Specifically, a comprehensive analysis is performed to obtain a predicted resource demand level assessment index for each user group. The specific analysis process is as follows: deploy a number of time monitoring points, divide the time monitoring points into historical time monitoring points and current time monitoring points according to a preset time scale, collect basic resource usage data of each user group at each historical time monitoring point and each current time monitoring point, extract the critical CPU usage rate, critical memory usage and critical disk space occupancy from the cloud server database, and obtain the historical basic resource demand assessment index and current basic resource demand assessment index of each user group after processing.

[0040] In a specific embodiment, the historical basic resource demand assessment index and the current basic resource demand assessment index of each user group are quantitative assessment data obtained by comprehensive analysis of the basic resource usage data of each user group including CPU usage, memory usage and disk space occupancy.

[0041] Based on the historical basic resource demand assessment index and the current basic resource demand assessment index of each user group, a comprehensive analysis is performed to obtain the predicted resource demand assessment index of each user group.

[0042] It should be explained that in this embodiment, the CPU usage rate refers to the CPU usage rate of the user group in different time periods, reflecting its computing power requirements. Memory usage refers to the amount of memory used by the user group, which helps evaluate its demand for storage resources. Disk space usage refers to the disk space occupied by the user group, including data storage, log files, etc. The CPU usage rate, memory usage and disk space usage can all be obtained from the historical data monitored in real time by a third-party system monitoring tool.

[0043] In a specific embodiment, the numerical expression of the predicted resource demand evaluation index of each user group is:

[0044] Y = ln[(tXn*ρ1+τV*P2)+1];

[0045] Where Y n represents the predicted resource demand evaluation index of the nth user group, n represents the number of each user group, n=1,2,3,...,m, m represents the total number of user groups, τX n represents the historical basic resource demand assessment index of the nth user group, s represents the number of each historical time monitoring point, s = 1, 2, 3, ..., h, h represents the total number of historical time monitoring points, Zs n Cs represents the CPU usage of the nth user group at the sth historical time monitoring point. nPs represents the memory usage of the nth user group at the sth historical time monitoring point. n represents the disk space usage of the nth user group at the sth historical time monitoring point, τV n Represents the current basic resource demand assessment index of the nth user group, i represents the number of each monitoring point at the current time, i=1,2,3,...,t, t represents the total number of monitoring points at the current time, Zi n represents the CPU usage of the nth user group at the i-th current time monitoring point, Ci n represents the memory usage of the nth user group at the i-th current time monitoring point, and Pin represents the disk space usage of the nth user group at the i-th current time monitoring point. Indicates the critical CPU usage, Indicates critical memory usage, represents the critical disk space occupancy, σ1 represents the historical basic resource demand assessment impact factor corresponding to the set CPU usage, σ2 represents the historical basic resource demand assessment impact factor corresponding to the set memory usage, σ3 represents the historical basic resource demand assessment impact factor corresponding to the set disk space occupancy, σ4 represents the current basic resource demand assessment impact factor corresponding to the set CPU usage, σ5 represents the current basic resource demand assessment impact factor corresponding to the set memory usage, σ6 represents the current basic resource demand assessment impact factor corresponding to the set critical disk space occupancy, ρ1 represents the predicted resource demand degree assessment impact factor corresponding to the preset historical basic resource demand assessment index, and ρ2 represents the predicted resource demand degree assessment impact factor corresponding to the preset current basic resource demand assessment index.

[0046] It should be understood that the predicted resource demand level assessment index of each user group is a quantitative assessment data obtained through a comprehensive analysis of the historical basic resource demand assessment index and the current basic resource demand assessment index of each user group, and is used to quantitatively assess the future demand level for basic resources of each user group, providing a basis for the rational allocation of cloud server resources.

[0047] It needs to be explained that Figure 2As shown, the predicted resource demand degree assessment impact factor corresponding to the historical basic resource demand assessment index is set to 0.6, and the predicted resource demand degree assessment impact factor corresponding to the current basic resource demand assessment index is set to 0.4. Curve a represents the relationship between the historical basic resource demand assessment index and the predicted resource demand degree assessment index when the current basic resource demand assessment index is 2.4. Curve b represents the relationship between the historical basic resource demand assessment index and the predicted resource demand degree assessment index when the current basic resource demand assessment index is 3.3. Curve c represents the relationship between the historical basic resource demand assessment index and the predicted resource demand degree assessment index when the current basic resource demand assessment index is 4.1.

[0048] It should be understood that, from the above embodiments, when the historical basic resource demand assessment index is larger and the current basic resource demand assessment index is also larger, the predicted resource demand assessment index is larger, indicating that the resource demand of the user group is larger.

[0049] It should be explained that in the above embodiment, the value range of the historical basic resource demand assessment impact factor corresponding to the CPU usage, the historical basic resource demand assessment impact factor corresponding to the memory usage, and the historical basic resource demand assessment impact factor corresponding to the disk space occupancy is [0, 1]. Through the relationship between the CPU usage, memory usage and disk space occupancy and the historical basic resource demand assessment index in the historical data, a mapping set of the CPU usage, memory usage and disk space occupancy and their corresponding historical basic resource demand assessment impact factors is established. The real-time CPU usage, memory usage and disk space occupancy are input to obtain the historical basic resource demand assessment impact factor corresponding to the CPU usage, the historical basic resource demand assessment impact factor corresponding to the memory usage, and the historical basic resource demand assessment impact factor corresponding to the disk space occupancy from the mapping set.

[0050] It should be explained that in the above embodiment, the value range of the current basic resource demand assessment impact factor corresponding to the CPU usage, the current basic resource demand assessment impact factor corresponding to the memory usage and the current basic resource demand assessment impact factor corresponding to the disk space occupancy is [0, 1]. Through the relationship between the CPU usage, memory usage and disk space occupancy and the current basic resource demand assessment index in the historical data, a mapping set of the CPU usage, memory usage and disk space occupancy and their corresponding current basic resource demand assessment impact factors is established. The real-time CPU usage, memory usage and disk space occupancy are input to obtain the current basic resource demand assessment impact factor corresponding to the CPU usage, the current basic resource demand assessment impact factor corresponding to the memory usage and the current basic resource demand assessment impact factor corresponding to the disk space occupancy from the mapping set.

[0051] It should be explained that in the above embodiment, the value range of the predicted resource demand degree assessment impact factor corresponding to the historical basic resource demand assessment index and the predicted resource demand degree assessment impact factor corresponding to the current basic resource demand assessment index is [0, 1]. Through the relationship between the historical basic resource demand assessment index and the current basic resource demand assessment index and the predicted resource demand degree assessment index in the historical data, a mapping set of the historical basic resource demand assessment index and the current basic resource demand assessment index and their corresponding predicted resource demand degree assessment impact factors is established. The real-time historical basic resource demand assessment index and the current basic resource demand assessment index are input to obtain the predicted resource demand degree assessment impact factor corresponding to the historical basic resource demand assessment index and the predicted resource demand degree assessment impact factor corresponding to the current basic resource demand assessment index from the mapping set.

[0052] It should be understood that all of the above-mentioned impact factors are extracted from the cloud server database.

[0053] It should be explained that the above embodiment can help the cloud platform automatically adjust resource allocation by comprehensively analyzing the current and historical basic resource demand indexes to obtain a predicted resource demand evaluation index, ensuring rapid expansion during peak demand and releasing excess resources during low demand, thereby achieving cost optimization. By predicting the degree of resource demand, redundant resources can be deployed in advance to ensure that services can still operate normally in the event of hardware failure or sudden traffic, thereby improving the high availability and disaster recovery capabilities of the system. At the same time, accurate demand forecasting and resource allocation help reduce unnecessary resource consumption and reduce energy costs.

[0054] Furthermore, resources are pre-allocated based on the predicted resource demand level assessment index of each user group. The specific allocation process is as follows: the predicted resource demand level assessment index of each user group is compared with the cloud server resource demand corresponding to each predicted resource demand level assessment index interval stored in the cloud server database to obtain the cloud server resource demand of each user group, and resources are allocated to each user group based on the cloud server resource demand of each user group.

[0055] It should be explained that, in a specific embodiment, a mapping set between the predicted resource demand degree assessment index interval and the cloud server resource demand is constructed based on historical data, the actual predicted resource demand degree assessment index is obtained, and then the interval to which the actual predicted resource demand degree assessment index belongs is determined, and the cloud server resource demand corresponding to the predicted resource demand degree assessment index interval is obtained from the mapping set.

[0056] It should be understood that the above embodiments can achieve resource isolation and guarantee. Each user group has its own resource pool, which means that they can operate independently within their own resource pools without being affected by other user groups. This provides resource isolation, ensuring the stability and security of resources. When the resource requirements of a certain user group suddenly increase, additional resources can be obtained from its reserved buffer resources without immediately affecting other user groups. In this way, the system can quickly respond to changes in resource requirements and improve the user experience. By pre-allocating resources, over-reservation and waste of resources can also be avoided. At the same time, the resource pool can ensure that each user group has sufficient resources to run its applications and services, avoiding performance degradation caused by resource contention. Equal-proportion division of resources can also ensure fairness among different user groups. Even in the case of limited resources, each user group can obtain corresponding resources according to its demand degree evaluation index. Predicting resource requirements and establishing resource pools helps to better plan future resource requirements, enabling the system to make adjustments in advance to cope with business growth or seasonal changes.

[0057] According to the business data, comprehensively analyze to obtain the resource allocation frequency reference index for each user group, monitor the load data of each user group, and comprehensively analyze to obtain the resource utilization degree evaluation index for each user group.

[0058] Specifically, comprehensively analyze to obtain the resource allocation frequency reference index for each user group. The specific analysis process is as follows: Collect the business data of each user group at each historical time monitoring point and each current time monitoring point, extract the critical transaction volume and critical access volume from the cloud server database, and obtain the historical business demand evaluation index and current business demand evaluation index for each user group after processing.

[0059] In a specific embodiment, the historical business demand evaluation index and current business demand evaluation index for each user group are quantitative evaluation data comprehensively analyzed based on the business data of each user group including transaction volume and access volume;

[0060] According to the historical business demand evaluation index and current business demand evaluation index of each user group, comprehensively analyze to obtain the resource allocation frequency reference index for each user group.

[0061] In a specific embodiment, the numerical expression of the resource allocation frequency reference index for each user group is:

[0062] W = tanh(τQn * 81 + TKn * ε2);

[0063] In the formula, W n represents the resource allocation frequency reference index of the nth user group, τQ n represents the historical business demand evaluation index of the nth user group, Msn Ns represents the transaction volume of the nth user group at the sth historical time monitoring point. n represents the number of visits of the nth user group at the sth historical time monitoring point, τK n represents the historical business demand assessment index of the nth user group, Min represents the transaction volume of the nth user group at the i-th current monitoring point, Ni n It represents the number of visits of the nth user group at the i-th current monitoring point. represents the critical trading volume, Indicates the critical number of visits, Indicates the historical business demand assessment impact factor corresponding to the preset transaction volume. Indicates the historical business demand assessment impact factor corresponding to the preset visit volume. Indicates the impact factor of the current business demand assessment corresponding to the preset transaction volume. It represents the current business demand assessment impact factor corresponding to the preset access volume, ε1 represents the resource allocation frequency impact factor corresponding to the preset historical business demand assessment index, and ε2 represents the resource allocation frequency impact factor corresponding to the preset current business demand assessment index.

[0064] It should be explained that, from the above embodiments, when the historical business demand assessment index and the current business demand assessment index of the user group are larger, the resource allocation frequency reference index of the user group is also larger, indicating that the resource allocation frequency of the user group is more frequent.

[0065] It should be understood that the resource allocation frequency reference index of each user group is a quantitative evaluation data obtained by comprehensively analyzing the business volume evaluation index of each user group at each time monitoring point in each historical monitoring period and the current monitoring period, which is used to quantitatively evaluate the degree of demand for future business of each user group and provide a basis for the reasonable allocation of cloud server resources.

[0066] It should be explained that the transaction volume and visit volume in this embodiment reflect the activity and resource demand of the user group at the business level. Through historical data analysis, the speed and pattern of business growth of the user group can be understood, and future resource demand can be predicted. The transaction volume and visit volume can be obtained by accessing the database of the internal backend statistical tool of the system.

[0067] It should be explained that in the above embodiment, the value range of the historical business demand assessment impact factor corresponding to the transaction volume and the historical business demand assessment impact factor corresponding to the visit volume is [0, 1]. Through the relationship between the transaction volume and visit volume in the historical data and the historical business demand assessment index, a mapping set of the transaction volume and visit volume and their corresponding historical business demand assessment impact factors is established, and the real-time transaction volume and visit volume are input to obtain the historical business demand assessment impact factor corresponding to the transaction volume and the historical business demand assessment impact factor corresponding to the visit volume from the mapping set.

[0068] It should be explained that in the above embodiment, the value range of the current business demand assessment impact factor corresponding to the transaction volume and the current business demand assessment impact factor corresponding to the visit volume is [0, 1]. Through the relationship between the transaction volume and visit volume in the historical data and the current business demand assessment index, a mapping set of the transaction volume and visit volume and their corresponding current business demand assessment impact factors is established, and the real-time transaction volume and visit volume are input to obtain the current business demand assessment impact factor corresponding to the transaction volume and the current business demand assessment impact factor corresponding to the visit volume from the mapping set.

[0069] It needs to be explained that in the above embodiment, the value range of the resource allocation frequency impact factor corresponding to the historical business demand assessment index and the resource allocation frequency impact factor corresponding to the current business demand assessment index is [0, 1]. Through the relationship between the historical business demand assessment index and the current business demand assessment index and the resource allocation frequency reference indicator in the historical data, a mapping set of the historical business demand assessment index and the current business demand assessment index and their corresponding resource allocation frequency impact factors is established. The real-time historical business demand assessment index and the current business demand assessment index are input to obtain the resource allocation frequency impact factor corresponding to the historical business demand assessment index and the resource allocation frequency impact factor corresponding to the current business demand assessment index from the mapping set.

[0070] It should be understood that all of the above-mentioned impact factors are extracted from the cloud server database.

[0071] It should be explained that the above embodiment obtains the resource allocation frequency reference index by comprehensively analyzing the current and historical business demand assessment index, which can more accurately identify the change pattern of resource demand, including peak, valley and average usage. This helps the cloud platform to finely adjust resource allocation, avoid resource waste and shortage, and ensure that resources are in place when needed. The resource allocation frequency reference index can guide the implementation of dynamic adjustment strategies. For example, when a rise in business demand is detected, the cloud platform can increase resource allocation in advance, otherwise it will be reduced. This immediate response mechanism can improve resource utilization efficiency while maintaining good service quality and user experience. Reasonable allocation of cloud server resources can effectively control costs. Based on historical and current demand indexes, it is possible to predict resource demand trends, optimize resource purchase and leasing strategies, avoid unnecessary expenses, and ensure that there are sufficient resources to support peak demand. By analyzing the resource demand frequency, it is possible to identify which time periods or business scenarios have higher resource demand. This helps to optimize applications and infrastructure, such as increasing cache, optimizing database queries, or upgrading hardware to improve overall system performance and response speed. The resource allocation frequency reference index helps to ensure business continuity and system reliability. When periods of high demand are predicted, additional resources can be pre-allocated to prevent service interruptions or performance degradation, ensuring that critical business processes run smoothly.

[0072] Furthermore, the comprehensive analysis obtains a resource utilization evaluation index for each user group. The specific analysis process is: deploying several time recording points, monitoring the load data of each user group in real time, wherein the load data includes the number of concurrent requests, response time and throughput of each user group at each time recording point, extracting the critical number of concurrent requests, critical response time and critical throughput from the cloud server database, and obtaining the resource utilization evaluation index for each user group through comprehensive analysis.

[0073] In a specific embodiment, the numerical expression of the resource utilization evaluation index of each user group is:

[0074]

[0075] Where, L n represents the resource utilization evaluation index of the nth user group, q represents the number of each time recording point, q=1,2,3,...,g, g represents the total number of time recording points, e represents a natural constant, α nq represents the number of concurrent requests of the nth user group at the qth time point, β nq represents the throughput of the nth user group at the qth time recording point, Indicates the critical number of concurrent requests. represents the critical throughput, θ1 represents the resource utilization assessment impact factor corresponding to the set number of concurrent requests, and θ2 represents the resource utilization assessment impact factor corresponding to the set throughput.

[0076] Table 1 Example of resource utilization evaluation index data

[0077]

[0078] It should be explained that, as shown in Table 1, in a specific embodiment, there are five groups of data, each group of data is the same, the critical number of concurrent requests is set to 100, the critical throughput is set to 1000, the resource utilization assessment impact factor corresponding to the concurrent request number is set to 0.5, and the resource utilization assessment impact factor corresponding to the throughput is set to 0.8.

[0079] It should be understood that, from the above embodiments, when the number of concurrent requests and the throughput of the user group are larger, the resource utilization evaluation index of the user group is larger, indicating that the higher the resource utilization rate, the more resources are needed.

[0080] It should be explained that in the above embodiment, the value range of the resource utilization assessment impact factor corresponding to the number of concurrent requests and the resource utilization assessment impact factor corresponding to the throughput is [0, 1]. Through the relationship between the number of concurrent requests and throughput and the resource utilization assessment index in historical data, a mapping set of the number of concurrent requests and throughput and their corresponding resource utilization assessment impact factors is established. The real-time number of concurrent requests and throughput are input to obtain the resource utilization assessment impact factor corresponding to the number of concurrent requests and the resource utilization assessment impact factor corresponding to the throughput from the mapping set. The impact factors are all extracted from the cloud server database.

[0081] It should be understood that the resource utilization evaluation index of each user group is a quantitative evaluation data obtained by comprehensively analyzing the number of concurrent requests, response time and throughput of each user group at each time recording point, which is used to quantitatively evaluate the resource utilization of each user group and provide a basis for the reasonable allocation of cloud server resources.

[0082] It should be explained that the number of concurrent requests in this embodiment reflects the load pressure of the system at the same time. When the number of requests increases, the system needs to allocate more resources to process these requests, resulting in an increase in resource utilization. Throughput reflects the ability of the system to process requests. A higher throughput usually means that the system can use resources more efficiently. However, when the throughput approaches or exceeds the system limit, resource utilization will rise sharply, which may cause system performance to degrade. Both the number of concurrent requests and throughput can be obtained by accessing the database of the system's internal background statistical tool.

[0083] It should be explained that in this embodiment, by monitoring and analyzing the utilization of resources, over-configuration of resources can be avoided and unnecessary expenses can be reduced. For example, if it is found that the CPU or memory utilization of some cloud servers is low within a specific time, the configuration can be adjusted appropriately to reduce resource consumption and reduce costs. Resource utilization analysis can support dynamic resource allocation strategies, that is, increase resources during peak demand periods and reduce resources during off-peak periods to ensure that resource usage matches actual demand and further optimize costs. In-depth analysis of resource utilization can help identify performance bottlenecks, such as excessive use of CPU, memory, storage or network resources, so as to take measures to optimize the system architecture and improve overall performance. Based on the analysis of resource utilization, the cloud platform can quickly respond to changes in business needs and automatically or manually adjust resources to cope with unforeseen traffic peaks or business growth. The elastic characteristics of cloud servers allow resources to be automatically expanded or contracted according to resource needs, ensuring that the best service experience can be provided at any point in time.

[0084] The resource pools of each user group are reallocated according to the resource allocation frequency reference index and resource utilization evaluation index of each user group.

[0085] Specifically, the resource pool of each user group is reallocated according to the resource allocation frequency reference index and resource utilization evaluation index of each user group. The specific allocation process is: comparing the resource allocation frequency reference index of each user group with the preset interval of the resource allocation frequency reference index threshold stored in the cloud server database. If the resource allocation frequency reference index of a user group is higher than the upper limit of the preset interval of the resource allocation frequency reference index or lower than the lower limit of the preset interval of the resource allocation frequency reference index, the user group is marked as a demand reallocation user group;

[0086] It should be explained that if the resource allocation frequency reference index of a user group is higher than the upper limit of the preset interval of the resource allocation frequency reference index or lower than the lower limit of the preset interval of the resource allocation frequency reference index, it indicates that the tasks of the user group are frequent or idle, and it is necessary to reallocate resources to the resource pool of the user group; if the resource allocation frequency reference index of a user group is within the preset interval of the resource allocation frequency reference index, it indicates that the task load of the user group is appropriate, and it is no longer necessary to reallocate resources to the resource pool of the user group;

[0087] Obtain resource utilization evaluation indexes of each demand redistribution user group, sort them in ascending order, and record the resource utilization evaluation index in the binary sequence as the reference standard resource utilization evaluation index;

[0088] The resource utilization evaluation index of each demand reallocation user group is subtracted from the resource utilization evaluation index of the reference standard, and the absolute value of the difference result is marked as the resource reallocation reference index of each demand reallocation user group. Resource reallocation is performed on each demand reallocation user group according to the resource reallocation reference index of each demand reallocation user group.

[0089] According to the matching of each difference interval, the lack or idle resource amount of each corresponding user group is obtained, and the idle resource amount is extracted from the resource pool of each user group and then allocated to the user group lacking resources.

[0090] In a specific embodiment, the cloud server database includes the critical CPU usage, critical memory usage and critical disk space occupancy, the predicted resource demand degree assessment impact factor corresponding to the preset historical basic resource demand assessment index and the predicted resource demand degree assessment impact factor corresponding to the preset current basic resource demand assessment index, and other data extracted from the above embodiments. The data of the cloud server database can be integrated into the cloud database through third-party Extract, Transform, Load tools and data warehouse services for data extraction, conversion and loading.

[0091] It should be explained that the reallocation process of the above embodiment can accurately identify changes in resource demand by comparing the resource allocation frequency reference index with the threshold, ensure that resources are allocated when they are really needed, and avoid resource waste. Based on the sorting and comparison of the resource utilization evaluation index, the resource use between each user group can be effectively balanced to ensure that the resource utilization efficiency is maximized and idle resources are reduced. The resource reallocation strategy can avoid over-configuration of user groups that do not require a large amount of resources, thereby saving costs. Dynamically adjusting resource allocation according to resource demand can provide more resources during peak periods, recover unused resources during trough periods, and avoid unnecessary expenditure of fixed costs. Timely replenishment of resources ensures that the system can respond quickly when demand is high, improving user experience and satisfaction. By maintaining a balance between resource supply and demand, the stability and consistency of services are ensured, and service degradation due to resource shortages is avoided. Reasonable allocation of resources and rapid reallocation capabilities help to quickly restore services, reduce downtime, and enhance business continuity. The resource reallocation mechanism allows flexible adjustment of resources according to real-time needs, supporting rapid expansion or contraction of services. It can quickly respond to changes in business needs and improve the overall agility and market response speed of the organization.

[0092] It needs to be explained that the predicted resource demand level assessment index, resource allocation frequency reference index and resource utilization level assessment index of each user group in the above embodiment are used for resource allocation. Resource pre-allocation is performed based on the predicted resource demand level assessment index of each user group. Based on the resource allocation frequency reference index of a certain user group, it is determined whether the user group needs reallocation. Finally, the amount of resources to be reallocated to each user group requiring reallocation is determined based on the resource utilization level assessment index of each user group. The three are in a hierarchical relationship.

[0093] The second aspect of the present invention provides a resource allocation management computer device for a cloud server, characterized in that it includes: a processor and a memory and a network interface connected to the processor; the network interface is connected to a non-volatile memory in the server; the processor calls a computer program from the non-volatile memory through the network interface during operation, and runs the computer program through the memory to execute any one of the methods described in the above inventions.

[0094] The above contents are merely examples and explanations of the structure of the present invention. The technicians in this technical field may make various modifications or additions to the specific embodiments described or replace them in a similar manner. As long as they do not deviate from the structure of the invention or exceed the scope defined by the claims, they should all fall within the protection scope of the present invention.

Claims

1. A resource allocation management method for a cloud server, characterized in that: include: Collect resource demand data of each user group, where the resource demand data includes basic resource usage data and business data; Based on the basic resource usage data, a comprehensive analysis is performed to obtain the predicted resource demand evaluation index of each user group, and resources are pre-allocated according to the predicted resource demand evaluation index of each user group, and the pre-allocated resources of each user group are marked as the resource pool of each user group; Based on the business data, a comprehensive analysis is conducted to obtain the resource allocation frequency reference index of each user group, and the load data of each user group is monitored and a comprehensive analysis is conducted to obtain the resource utilization evaluation index of each user group; Re-allocate resources in the resource pool of each user group according to the resource allocation frequency reference index and resource utilization evaluation index of each user group; The predicted resource demand evaluation index of each user group. The specific analysis process is as follows: Deploy several time monitoring points, divide the several time monitoring points into historical time monitoring points and current time monitoring points according to the preset time scale, collect basic resource usage data of each user group at each historical time monitoring point and each current time monitoring point, extract the critical CPU usage rate, critical memory usage and critical disk space occupancy from the cloud server database, and obtain the historical basic resource demand assessment index and current basic resource demand assessment index of each user group after processing; Based on the historical basic resource demand assessment index and the current basic resource demand assessment index of each user group, a comprehensive analysis is conducted to obtain the predicted resource demand assessment index of each user group; The comprehensive analysis obtains the resource allocation frequency reference index of each user group. The specific analysis process is as follows: Collect the business data of each user group at each historical time monitoring point and each current time monitoring point, extract the critical transaction volume and critical access volume from the cloud server database, and obtain the historical business demand assessment index and current business demand assessment index of each user group after processing; Based on the historical business demand assessment index and current business demand assessment index of each user group, a comprehensive analysis is performed to obtain a reference index for resource allocation frequency of each user group; The comprehensive analysis obtains the resource utilization evaluation index of each user group. The specific analysis process is as follows: Deploy several time recording points to monitor the load data of each user group in real time. The load data includes the number of concurrent requests, response time and throughput of each user group at each time recording point. Extract the critical number of concurrent requests, critical response time and critical throughput from the cloud server database, and comprehensively analyze to obtain the resource utilization evaluation index of each user group. The resource pool of each user group is redistributed according to the resource allocation frequency reference index of each user group and the resource utilization evaluation index. The specific allocation process is as follows: The resource allocation frequency reference index of each user group is compared with the preset interval of the resource allocation frequency reference index threshold stored in the cloud server database. If the resource allocation frequency reference index of a user group is higher than the upper limit of the preset interval of the resource allocation frequency reference index or lower than the lower limit of the preset interval of the resource allocation frequency reference index, the user group is marked as a demand redistribution user group; Obtain resource utilization evaluation indexes of each demand redistribution user group, sort them in ascending order, and record the resource utilization evaluation index in the binary sequence as the reference standard resource utilization evaluation index; Subtract the resource utilization evaluation index of each demand reallocation user group from the reference standard resource utilization evaluation index, and mark the absolute value of the difference result as the resource reallocation reference index of each demand reallocation user group, and perform resource reallocation for each demand reallocation user group according to the resource reallocation reference index of each demand reallocation user group; The numerical expression of the resource allocation frequency reference index of each user group is: ； In the formula, It represents the resource allocation frequency reference index of the nth user group. represents the historical business demand assessment index of the nth user group, , represents the transaction volume of the nth user group at the sth historical time monitoring point, It represents the number of visits of the nth user group at the sth historical time monitoring point, where s represents the number of each historical time monitoring point. , h represents the total number of historical time monitoring points, represents the historical business demand assessment index of the nth user group, , represents the transaction volume of the nth user group at the i-th current monitoring point, It represents the number of visits of the nth user group at the i-th current time monitoring point, where i represents the number of each current time monitoring point. , t represents the total number of monitoring points at the current time, represents the critical trading volume, Indicates the critical number of visits, Indicates the historical business demand assessment impact factor corresponding to the preset transaction volume. Indicates the historical business demand assessment impact factor corresponding to the preset visit volume. Indicates the impact factor of the current business demand assessment corresponding to the preset transaction volume. Indicates the impact factor of the current business demand assessment corresponding to the preset visit volume. Indicates the resource allocation frequency impact factor corresponding to the preset historical business demand assessment index. Indicates the resource allocation frequency influencing factor corresponding to the preset current business demand assessment index; The numerical expression of the resource utilization evaluation index of each user group is: ； In the formula, represents the resource utilization evaluation index of the nth user group, q represents the number of each time recording point, , g represents the total number of time recording points, e represents the natural constant, Indicates the number of concurrent requests of the nth user group at the qth time record point, represents the throughput of the nth user group at the qth time recording point, Indicates the critical number of concurrent requests. represents the critical throughput, Indicates the resource utilization evaluation impact factor corresponding to the set number of concurrent requests. Indicates the resource utilization evaluation impact factor corresponding to the set throughput; The numerical expression of the predicted resource demand evaluation index of each user group is: ； In the formula, It represents the predicted resource demand evaluation index of the nth user group, where n represents the number of each user group. , m represents the total number of user groups, represents the historical basic resource demand assessment index of the nth user group, , s represents the number of each historical time monitoring point, , h represents the total number of historical time monitoring points, represents the CPU usage of the nth user group at the sth historical time monitoring point, Indicates the memory usage of the nth user group at the sth historical time monitoring point, Indicates the disk space usage of the nth user group at the sth historical time monitoring point, Represents the current basic resource demand assessment index of the nth user group, , i represents the number of each monitoring point at the current time, , t represents the total number of monitoring points at the current time, represents the CPU usage of the nth user group at the i-th current time monitoring point, Indicates the memory usage of the nth user group at the i-th current time monitoring point, Indicates the disk space usage of the nth user group at the i-th current monitoring point. Indicates the critical CPU usage, Indicates critical memory usage, Indicates the critical disk space usage. Indicates the historical basic resource demand assessment impact factor corresponding to the set CPU usage rate, Indicates the historical basic resource demand assessment impact factor corresponding to the set memory usage. Indicates the historical basic resource demand assessment impact factor corresponding to the set disk space usage. Indicates the current basic resource demand assessment impact factor corresponding to the set CPU usage rate, Indicates the impact factor of the current basic resource demand assessment corresponding to the set memory usage. Indicates the impact factor of the current basic resource demand assessment corresponding to the set critical disk space usage. Indicates the impact factor of the predicted resource demand assessment corresponding to the preset historical basic resource demand assessment index, It indicates the predicted resource demand assessment impact factor corresponding to the preset current basic resource demand assessment index.

2. A cloud server resource allocation management method according to claim 1, characterized in that: The resource pre-allocation is performed according to the predicted resource demand evaluation index of each user group. The specific allocation process is as follows: The predicted resource demand level assessment index of each user group is compared with the cloud server resource demand corresponding to each predicted resource demand level assessment index interval stored in the cloud server database to obtain the cloud server resource demand of each user group, and resources are allocated to each user group according to the cloud server resource demand of each user group.

3. The resource allocation management method of a cloud server according to claim 1, characterized in that: The resource utilization evaluation index of each user group is a quantitative evaluation data obtained by comprehensively analyzing the number of concurrent requests, response time and throughput of each user group at each time recording point, which is used to quantitatively evaluate the resource utilization of each user group and provide a basis for the reasonable allocation of cloud server resources.

4. A resource allocation management computer device for a cloud server, characterized in that: include: Processors and memory and network interfaces connected to the processors; The network interface is connected to a non-volatile memory in the server; When running, the processor retrieves a computer program from the non-volatile memory through the network interface, and runs the computer program through the memory to execute the method described in any one of claims 1 to 3.

Images