Cloud computing resource management method, system, device, medium and product
By dividing cloud computing resources into management sub-regions, collecting multi-dimensional parameters, and calculating the comprehensive resource optimization index, the problems of untimely and inefficient resource scheduling in traditional cloud computing resource management methods are solved, and refined dynamic management and efficient utilization of resources are realized.
Patent Information
- Application Number
- CN202511417744.8
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2025-09-30
- Publication Date
- 2026-01-09
AI Technical Summary
Traditional cloud computing resource management methods cannot achieve intelligent and refined resource management optimization, and are difficult to adapt to dynamic changes in business load. This results in untimely resource scheduling, slow response speed, inability to adapt to differentiated needs and hardware diversity in multiple scenarios, and problems of resource competition and low scheduling efficiency.
By dividing the target resource into management sub-regions, collecting multi-dimensional computing power demand information parameters, and calculating the comprehensive optimization index of computing resources, we can achieve refined and dynamic management of resources at the sub-region level, including the collection and comprehensive optimization of processor operating parameters, disk operating parameters, throughput parameters, and cost parameters.
It enables refined and dynamic resource management, improves the accuracy and efficiency of resource allocation, can respond promptly to changes in business load, rationally flow and allocate resources, adapt to multi-scenario needs, and reduce costs.
Smart Images
Figure CN121309505A_ABST
Abstract
Description
Technical Field
[0001] This invention relates to the field of cloud computing technology, and in particular to a cloud computing resource management method, system, device, medium and product. Background Technology
[0002] Traditional cloud computing resource management is based on a stable architecture and relies on predefined rules and policies to achieve resource scheduling. This model ensures the stability and reliability of resource allocation and sets clear boundaries for resource allocation to prevent overuse or abuse.
[0003] Based on pre-configured rules and measurements, some cloud computing platforms have further developed resource monitoring and auto-scaling functions. However, most of these functions are based on simple threshold rules or historical data calculations, which can only achieve mechanical resource adjustments and cannot achieve intelligent and refined resource management optimization. Summary of the Invention
[0004] The embodiments of the present invention aim to provide a cloud computing resource management method, system, device, medium and product, which breaks the traditional extensive overall resource management model, can focus on the computing power characteristics of the target main resource, and realize refined dynamic management of resources at the sub-region level.
[0005] This invention provides a cloud computing resource management method, including:
[0006] Acquire the target entity resources, and divide the target management area where the target entity resources are located into several management sub-areas;
[0007] Collect computing power demand information parameters for each management sub-region; the computing power demand information parameters include processor operating parameters, disk operating parameters, throughput parameters, and cost parameters;
[0008] Based on the computing power demand information parameters, calculate the comprehensive resource optimization index for each management sub-region;
[0009] Based on the aforementioned comprehensive resource optimization index, resources are allocated and managed among the sub-regions.
[0010] As an improvement to the above solution, the step of acquiring the target entity resource involves dividing the target management area where the target entity resource is located into several management sub-areas, including:
[0011] Acquire the target entity's resources and determine the area where the target entity's resources are located as the target management area;
[0012] Based on the logical attributes and operational requirements of the computing nodes in the target entity resources, the target entity resources are grouped.
[0013] Based on the grouping results of the target entity resources, the target management area is divided into several management sub-areas; each management sub-area includes at least one computing node.
[0014] As an improvement to the above scheme, the collection of computing power demand information parameters for each management sub-region includes:
[0015] The temperature and humidity of each management sub-area where the processor is running are collected to obtain the processor's operating parameters;
[0016] Collect the disk read speed, write speed, and operations per second of each managed sub-region to obtain disk operating parameters;
[0017] Collect the storage throughput and computing throughput of each management sub-region to obtain throughput parameters;
[0018] Collect the profit, cost, and investment amount of each management sub-region to obtain cost parameters;
[0019] The computing power requirement information parameters are obtained based on the processor operating parameters, the disk operating parameters, the throughput parameters, and the cost parameters.
[0020] As an improvement to the above scheme, the step of calculating the comprehensive resource optimization index of each management sub-region based on the computing power demand information parameters includes:
[0021] Based on the processor operating parameters, the rate deviation of each management sub-region is calculated, and the rate deviation is used as a standard normal distribution variable to obtain the processor rate index.
[0022] Based on the disk operating parameters, the disk utilization rate of each management sub-region is calculated, and based on the non-linear growth relationship between the disk utilization rate and disk efficiency, the disk efficiency index is obtained.
[0023] Based on the throughput parameters, the collaborative relationship between the throughput of each management sub-region and time correction is calculated to obtain the performance index;
[0024] Based on the cost parameters, the cost-effectiveness of each management sub-region is calculated to obtain the cost index;
[0025] Based on the processor speed index, the disk efficiency index, the performance index, and the cost index, the comprehensive resource optimization index for each management sub-region is obtained.
[0026] As an improvement to the above scheme, the processor operating parameters include the temperature and humidity of each management sub-region where the processor is in operation; then, the step of calculating the rate deviation of each management sub-region based on the processor operating parameters, and using the rate deviation as a standard normal distribution variable to obtain the processor rate index, includes:
[0027] The total temperature of all management sub-regions is obtained based on the temperature of each management sub-region where the processor is in operation.
[0028] Based on the temperature, the total temperature, and the total processor running time, the difference between the temperature of each management sub-region and the total temperature is calculated to obtain the rate temperature deviation.
[0029] The average humidity of each management sub-region is obtained based on the humidity of the processors in operation within each management sub-region.
[0030] Based on the humidity and the average humidity, the difference between the humidity of each managed sub-area and the average humidity is calculated to obtain the rate humidity deviation.
[0031] The rate deviation is obtained by multiplying the rate-temperature deviation and the rate-humidity deviation.
[0032] The processor rate index is calculated by treating the rate deviation as a standard normal distribution variable.
[0033] As an improvement to the above scheme, the disk operating parameters include disk read speed, write speed, and operations per second; then, based on the disk operating parameters, the disk utilization of each management sub-region is calculated, and based on the non-linear growth relationship between disk utilization and disk efficiency, a disk efficiency index is obtained, including:
[0034] The sum of the read speed and write speed of each management sub-region is calculated to obtain the total disk transfer rate;
[0035] The quotient of the sum of the disk transfer rates and the number of operations per second is used as the disk utilization rate;
[0036] The disk utilization rate is incremented by 1 and then logarithmically calculated to obtain the disk efficiency index.
[0037] As an improvement to the above scheme, the throughput parameters include storage throughput and computing throughput; then, based on the throughput parameters, the collaborative relationship of throughput of each management sub-region adjusted over time is calculated to obtain a performance index, including:
[0038] The average storage throughput is calculated based on the storage throughput and total time of each management sub-region;
[0039] The average computing throughput is calculated based on the computing throughput and total time of each management sub-region.
[0040] Based on the storage throughput, the average storage throughput, the computing throughput, and the average computing throughput, a relative synergistic relationship between computing throughput and storage throughput is obtained.
[0041] The performance index is derived based on the relative synergistic relationship between computation throughput and storage throughput.
[0042] As an improvement to the above scheme, the cost parameters include profit, cost, and investment amount; then, the step of calculating the cost-effectiveness of each management sub-region based on the cost parameters to obtain a cost index includes:
[0043] Calculate the difference between the profit and the cost for each management sub-region to obtain the net profit increment;
[0044] Calculate the quotient of the net profit increment and the investment amount to obtain the cost-effectiveness of each management sub-region;
[0045] Based on the aforementioned cost-effectiveness, the cost index for each management sub-region is obtained.
[0046] As an improvement to the above scheme, after calculating the comprehensive resource optimization index of each management sub-region based on the computing power demand information parameters, the cloud computing resource management method further includes:
[0047] Based on the comprehensive resource optimization index and the preset early warning threshold, it is determined whether there is any abnormal usage of cloud computing resources, and an early warning signal is sent when abnormal usage occurs.
[0048] As an improvement to the above scheme, the allocation and management of resources among the management sub-regions based on the comprehensive resource optimization index includes:
[0049] If the comprehensive resource optimization index of the managed sub-region is less than the preset optimization index threshold, the managed sub-region will be regarded as a sub-region to be optimized.
[0050] A resource optimization strategy is formulated that includes the sub-region to be optimized, and the resource optimization strategy is sent to each management sub-region to allocate and manage resources among the management sub-regions; the resource optimization strategy includes the addition, reduction, migration and load balancing of cloud computing resources.
[0051] This invention also provides a cloud computing resource management system, comprising:
[0052] The sub-region division module is used to acquire target entity resources and divide the target management area where the target entity resources are located into several management sub-regions.
[0053] The parameter acquisition module is used to collect computing power demand information parameters for each management sub-region; the computing power demand information parameters include processor operating parameters, disk operating parameters, throughput parameters, and cost parameters;
[0054] The index calculation module is used to calculate the comprehensive resource optimization index of each management sub-region based on the computing power demand information parameters.
[0055] The resource optimization module is used to allocate and manage resources among the managed sub-regions based on the comprehensive resource optimization index.
[0056] This invention also provides a cloud computing resource management device, including a processor, a memory, and a computer program stored in the memory and configured to be executed by the processor. When the processor executes the computer program, it implements the cloud computing resource management method described above.
[0057] This invention also provides a computer-readable storage medium, which includes a stored computer program, wherein the computer program, when running, controls the device where the computer-readable storage medium is located to execute the cloud computing resource management method described above.
[0058] This invention also provides a computer program product, which includes a computer program or computer instructions. When the computer program or computer instructions are executed by a processor, they perform the cloud computing resource management method described above.
[0059] Compared with existing technologies, the present invention discloses a cloud computing resource management method, system, device, medium, and product. This method involves acquiring target resources and dividing the target management area containing those resources into several management sub-regions; collecting computing power demand information parameters for each sub-region; these parameters include processor operating parameters, disk operating parameters, throughput parameters, and cost parameters; calculating a comprehensive resource optimization index for each sub-region based on these parameters; and allocating and managing resources among the sub-regions according to the comprehensive resource optimization index. By employing this invention, the computing power characteristics of the target resources can be focused on, enabling refined and dynamic resource management at the sub-region level. Attached Figure Description
[0060] Figure 1 This is a flowchart illustrating the steps of a cloud computing resource management method provided in an embodiment of the present invention;
[0061] Figure 2 This is a schematic diagram of the structure of a cloud computing resource management system provided in an embodiment of the present invention;
[0062] Figure 3 A schematic diagram of the structure of a cloud computing resource management device provided in an embodiment of the present invention. Detailed Implementation
[0063] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.
[0064] In the description and claims, it should be understood that the terms "first," "second," etc., used in the description and claims are only for the purpose of distinguishing the description of the same technical features, and should not be construed as indicating or implying relative importance or implicitly specifying the number of technical features indicated, nor necessarily the order of description or chronological order. The terms are interchangeable where appropriate. Therefore, a feature defined as "first" or "second" may explicitly or implicitly include at least one of that feature.
[0065] Traditional cloud computing resource management is typically built on a stable architecture, relying on predefined rules and policies. However, pre-configuration methods struggle to accurately adapt to dynamic changes in business load due to a lack of deep understanding of computing power demands. This often leads to problems such as untimely resource scheduling and slow response times during sudden business peaks or when handling complex computing tasks. Furthermore, existing cloud computing resource management methods prioritize stable resource operation, making it difficult to adapt to the differentiated needs of various scenarios, hardware diversity, and the complexity of global collaboration. This results in limitations when facing challenges such as resource contention in multi-tenant environments, efficient utilization of heterogeneous hardware resources, and cross-regional resource scheduling.
[0066] Based on the above considerations, this invention provides a cloud computing resource management method. Please refer to... Figure 1 In this embodiment, the cloud computing resource management method is specifically executed through steps S1 to S4:
[0067] S1. Obtain the target entity resources and divide the target management area where the target entity resources are located into several management sub-areas.
[0068] It should be noted that the target resource can be any cloud computing resource owned and managed by any organization or entity. This target resource can originate from different organizations or entities, as long as their cloud computing resources can be scheduled and managed together. In other words, the embodiments of the present invention can flexibly adapt to various scenarios, such as a single scenario like a physical server room in a data center, as well as complex scenarios such as multi-tenant scenarios, cross-geographical region scenarios, and hybrid scenarios.
[0069] In the above scheme, the division of management sub-regions revolves around the target resources that need to be managed, avoiding the problem of a large number of invalid resources being included in the fixed management scope due to the passive inclusion of resources in the traditional scheme, which increases redundant operations and improves management accuracy.
[0070] S2. Collect computing power demand information parameters for each management sub-region.
[0071] In some preferred embodiments, the computing power requirement information parameters include processor operating parameters, disk operating parameters, throughput parameters, and cost parameters.
[0072] Existing cloud computing resource management methods generally rely on static threshold rules, which only scratch the surface of computing power demand and cannot capture sudden changes in business load. They also have shortcomings in identifying differences in load characteristics.
[0073] In this embodiment of the invention, by collecting multi-dimensional computing power demand information parameters such as processor operating parameters, disk operating parameters, throughput parameters, and cost parameters, it is possible to capture the degree of computing power saturation, storage resource shortage, and data interaction capability in cloud computing, measure the economics of resource use, and further accurately identify the computing power of each management sub-region to avoid resource mismatch.
[0074] S3. Calculate the comprehensive resource optimization index for each management sub-region based on the computing power demand information parameters.
[0075] By converting multi-dimensional, dynamically changing computing power demand information parameters into a quantifiable comprehensive resource optimization index, the problem of inefficiency in fragmented decision-making based on a single parameter in existing technologies is solved, enabling a resource management solution that shifts from passive response to proactive optimization based on computing power.
[0076] S4. Based on the comprehensive resource optimization index, allocate and manage resources among the management sub-regions.
[0077] The comprehensive resource optimization index is calculated based on the computing power demand information parameters of each management sub-region. Using this index as a standard for resource allocation avoids blind allocation. Furthermore, by using the comprehensive resource optimization index of each management sub-region, the current resource allocation and usage status of the management sub-region can be clearly understood, improving the efficiency of cloud computing resource flow within the target management area.
[0078] In the above scheme, by dividing the management sub-regions based on the target entity's resources, a foundation is laid for subsequent refined management. By collecting multi-dimensional computing power demand information parameters, the resource operation and demand status of each management sub-region can be fully grasped. This transforms the complex resource status of each sub-region into a quantitative resource comprehensive optimization index that can be directly used for decision-making, promoting the rational flow and allocation of resources among the sub-regions. It can focus on the computing power characteristics of the target entity's resources and achieve refined dynamic management of resources at the sub-region level.
[0079] As a preferred implementation, step S1 involves acquiring the target entity resource and dividing the target management area where the target entity resource is located into several management sub-areas, including:
[0080] Acquire the target entity's resources and determine the area where the target entity's resources are located as the target management area;
[0081] Based on the logical attributes and operational requirements of the computing nodes in the target entity resources, the target entity resources are grouped.
[0082] Based on the grouping results of the target entity resources, the target management area is divided into several management sub-areas; each management sub-area includes at least one computing node.
[0083] It should be noted that the logical attributes refer to the inherent characteristics of the computing node at the logical level, reflecting the logical association attributes of the computing node; the operational requirements refer to the specific resource requirements of the computing node during operation, reflecting the operational performance and resource dependencies of the computing node.
[0084] In some preferred embodiments, the target resources can also be grouped according to other partitioning methods of computing nodes in a cloud computing environment. For example, they can be grouped according to the physical deployment of computing nodes. The grouped target resources have certain homogeneous characteristics, and the specific grouping rules do not affect the beneficial effects produced by the embodiments of the present invention.
[0085] It should also be noted that, in this embodiment of the invention, the computing node is a node with independent computing capabilities, including a processor. The processor can be a CPU, GPU, FPGA, etc., and the specific form of the processor does not affect the beneficial effects of this invention.
[0086] As a preferred implementation, step S2 involves collecting computing power demand information parameters for each management sub-region, including:
[0087] The temperature and humidity of each management sub-area where the processor is running are collected to obtain the processor's operating parameters;
[0088] Collect the disk read speed, write speed, and operations per second of each managed sub-region to obtain disk operating parameters;
[0089] Collect the storage throughput and computing throughput of each management sub-region to obtain throughput parameters;
[0090] Collect the profit, cost, and investment amount of each management sub-region to obtain cost parameters;
[0091] The computing power requirement information parameters are obtained based on the processor operating parameters, the disk operating parameters, the throughput parameters, and the cost parameters.
[0092] Temperature and humidity are critical environmental indicators for stable processor operation. Excessive temperature can lead to processor throttling, reduced computing power, or even burnout; excessive humidity may cause short circuits. The temperature and humidity of each management sub-region directly reflect the processor's health status and actual computing power output, and are fundamental parameters for measuring processor performance.
[0093] The disk is the core component for data storage and interaction. The read and write speed of the disk determines the data transfer efficiency, and the number of operations per second reflects the disk's ability to process requests.
[0094] Storage throughput reflects the speed at which a storage system processes read and write requests, expressed as bytes per second (MB / s) or GB / s, or I / O operations per second (IOPS); computing throughput measures the ability of a cloud server cluster or a single server to handle workloads.
[0095] In some preferred embodiments, the temperature and humidity of each management sub-region are obtained by collecting environmental data from the computing nodes where the key processors are located; in other preferred embodiments, the temperature and humidity of each management sub-region are obtained by collecting environmental data from the processors of all computing nodes and then weighting the data.
[0096] As a preferred implementation, step S3, calculating the comprehensive resource optimization index for each management sub-region based on the computing power demand information parameters, is executed through steps S31-S35:
[0097] S31. Based on the processor operating parameters, calculate the rate deviation of each management sub-region, and use the rate deviation as a standard normal distribution variable to obtain the processor rate index.
[0098] Processor speed deviation is affected by random factors such as hardware stability and load fluctuations, and its distribution usually approximates a normal distribution. In this embodiment of the invention, after converting the speed deviation into a standard normal distribution, the probability properties of the normal distribution can be directly used to determine the degree of abnormality of the deviation. Compared with the traditional method of determining speed using a speed threshold, this embodiment of the invention takes into account the fluctuation characteristics of the processor.
[0099] S32. Calculate the disk utilization rate of each management sub-region based on the disk operating parameters, and obtain the disk efficiency index based on the non-linear growth relationship between disk utilization rate and disk efficiency.
[0100] It's important to note that due to the physical limitations of disks, disk efficiency does not increase at a uniform rate with increasing utilization. When disk utilization is low, new read / write requests can be processed quickly; however, when disk utilization is high, request queuing intensifies to process new read / write requests. In other words, the higher the disk utilization, the more significant the decrease in the rate of efficiency improvement.
[0101] S33. Based on the throughput parameters, calculate the collaborative relationship between the throughput of each management sub-region and the time correction to obtain the performance index.
[0102] In this embodiment of the invention, the actual computing power of the managed sub-region is accurately determined by analyzing the matching degree and mutual influence between storage and computing throughput to identify their collaborative efficiency.
[0103] S34. Based on the cost parameters, calculate the cost-effectiveness of each management sub-region to obtain the cost index.
[0104] The business scale and resource base of different management sub-regions may vary greatly, and directly comparing absolute values is not very meaningful. This invention provides a unified economic benchmark for the allocation of computing resources in different management sub-regions by calculating cost-effectiveness.
[0105] S35. Based on the processor speed index, the disk efficiency index, the performance index, and the cost index, obtain the comprehensive resource optimization index for each management sub-region.
[0106] The resource optimization index integrates processor speed index, disk efficiency index, performance index, and cost index to reflect the adaptability of resource allocation in the managed sub-region to real-time business load, technology iteration, and cost fluctuations.
[0107] In some preferred embodiments, when performing step S3, the computing power demand information parameters are first linearly standardized, and then each index is calculated.
[0108] For example, the standardized computing power demand information parameter X is represented as:
[0109]
[0110] Where X0 represents the initial collected computing power demand information parameter, X min X is the minimum value of the initially collected computing power demand information parameters. max This represents the maximum value of the computing power demand information parameters initially collected.
[0111] Further, preferably, the processor operating parameters include the temperature and humidity of each management sub-region where the processor is in operation. Then, step S31, calculating the rate deviation of each management sub-region based on the processor operating parameters, and using the rate deviation as a standard normal distribution variable to obtain the processor rate index, includes:
[0112] The total temperature of all management sub-regions is obtained based on the temperature of each management sub-region where the processor is in operation.
[0113] Based on the temperature, the total temperature, and the total processor running time, the difference between the temperature of each management sub-region and the total temperature is calculated to obtain the rate temperature deviation.
[0114] The average humidity of each management sub-region is obtained based on the humidity of the processors in operation within each management sub-region.
[0115] Based on the humidity and the average humidity, the difference between the humidity of each managed sub-area and the average humidity is calculated to obtain the rate humidity deviation.
[0116] The rate deviation is obtained by multiplying the rate-temperature deviation and the rate-humidity deviation.
[0117] The processor rate index is calculated by treating the rate deviation as a standard normal distribution variable.
[0118] In some preferred embodiments, the rate-temperature deviation CR of the i-th management sub-region Ti Represented as:
[0119]
[0120] Among them, C Ti Let be the temperature of the i-th management sub-region, CT be the total temperature of all management sub-regions, Δt1 be the total processor running time, and α1 and α2 be preset temperature deviation adjustment coefficients.
[0121] In the above scheme, the degree of temperature deviation can be highlighted by squaring the difference between the temperature of the managed sub-region and the average temperature during the total processing time; the absolute level of temperature deviation can be reflected by calculating the absolute value of the difference between the temperature of the managed sub-region and the average temperature during the total processing time. This embodiment of the invention considers the impact of the current temperature difference on the rate temperature deviation from two dimensions.
[0122] Preferably, the rate humidity deviation CR of the i-th management sub-region Hi Represented as:
[0123]
[0124] Among them, H i Let the humidity be the value of the i-th managed sub-region. β represents the average humidity of each managed sub-region, β is the preset humidity deviation adjustment coefficient, and n is the number of managed sub-regions.
[0125] In the above scheme, the sum of the differences between the humidity of each management sub-region and the average humidity is first calculated to comprehensively consider the overall situation of each management sub-region deviating from the average humidity; then the summation result is normalized to reflect the dispersion of the humidity difference; finally, the logarithm is taken to weaken the extreme influence of large humidity deviations to a certain extent, so that the calculated rate humidity deviation result has high stability and can evenly reflect the contribution of humidity deviation of each management sub-region to rate humidity deviation.
[0126] Preferably, the processor rate index RU of the i-th management sub-region i Represented as:
[0127]
[0128] Among them, CR Ti For the rate-temperature deviation of the i-th management sub-region, CR Hi λ represents the rate-humidity deviation of the i-th management sub-region, and λ is the preset processor rate adjustment coefficient.
[0129] Calculating the processor speed index can help the system identify situations of insufficient or excessive resource allocation, thereby adjusting resource configuration in a timely manner to ensure that various tasks can be completed efficiently.
[0130] For example, in project management, by calculating resource utilization, project managers can reallocate workloads to prevent team members from being overworked or resources from being idle. In scenarios such as cloud computing, enterprises can pay based on the actual amount of resources used, avoiding the problems of large investment and maintenance costs required in traditional models. By monitoring resource utilization in real time, enterprises can more accurately control resource consumption and further reduce costs. As business grows, the demand for resources will continue to increase. By calculating resource utilization metrics, enterprises can predict future resource needs and prepare in advance, such as expanding computing resources and adding storage devices, to ensure the smooth progress of business growth.
[0131] Preferably, the disk operating parameters include the disk read speed, write speed, and operations per second; step S32: calculate the disk utilization rate of each management sub-region based on the disk operating parameters, and obtain the disk efficiency index based on the non-linear growth relationship between disk utilization and disk efficiency, including:
[0132] The sum of the read speed and write speed of each management sub-region is calculated to obtain the total disk transfer rate;
[0133] The quotient of the sum of the disk transfer rates and the number of operations per second is used as the disk utilization rate;
[0134] The disk utilization rate is incremented by 1 and then logarithmically calculated to obtain the disk efficiency index.
[0135] In some preferred embodiments, the disk utilization DU of the i-th management sub-region i Represented as:
[0136]
[0137] Among them, DRv i DWv represents the read speed of the i-th management sub-region. i For the write speed of the i-th management sub-region, DO i Let be the number of operations per second for the i-th management sub-region.
[0138] Furthermore, preferably, the disk efficiency index DE of the i-th management sub-region i Represented as:
[0139] DE i =η×ln(DU) i +1) (6);
[0140] Where η is a preset disk efficiency adjustment coefficient.
[0141] In the above scheme, adding 1 to the disk utilization rate and using it as the independent variable of the logarithmic function ensures that the independent variable is greater than 0, allowing the function to be calculated correctly. Simultaneously, when the disk is not in use, DU... i =0, at which point the corresponding logarithmic function has a value of 1, setting the zero point for efficiency calculation in the initial state of the disk.
[0142] By calculating disk efficiency metrics, enterprises can more rationally plan the use of storage resources. For example, for I / O-intensive applications, higher-performance disks or technologies such as RAID can be configured to improve overall performance; while for data-intensive applications, storage capacity can be increased to meet demand. Disk efficiency metrics provide strong data support for enterprise decision-making. Whether making decisions about hardware upgrades, system optimization, or business expansion, enterprises can use these metrics to evaluate the feasibility and effectiveness of different solutions.
[0143] Preferably, the throughput parameters include storage throughput and computing throughput; then step S33, based on the throughput parameters, calculates the collaborative relationship of throughput of each management sub-region over time to obtain a performance index, including:
[0144] The average storage throughput is calculated based on the storage throughput and total time of each management sub-region;
[0145] The average computing throughput is calculated based on the computing throughput and total time of each management sub-region.
[0146] Based on the storage throughput, the average storage throughput, the computing throughput, and the average computing throughput, a relative synergistic relationship between computing throughput and storage throughput is obtained.
[0147] The performance index is derived based on the relative synergistic relationship between computation throughput and storage throughput.
[0148] In some preferred embodiments, the performance index TI of the i-th management sub-region i Represented as:
[0149]
[0150] Among them, St i Let i be the storage throughput of the i-th management sub-region. Δt2 is the total time, Jt i Let i be the computational throughput of the i-th management sub-region. This is the preset performance adjustment coefficient.
[0151] By calculating performance index coefficients, businesses can identify and optimize inefficient processes, reduce unnecessary waste and waiting time, and improve overall operational efficiency.
[0152] Preferably, the cost parameters include profit, cost, and investment amount; then step S34, calculating the cost-effectiveness of each management sub-region based on the cost parameters to obtain a cost index, includes:
[0153] Calculate the difference between the profit and the cost for each management sub-region to obtain the net profit increment;
[0154] Calculate the quotient of the net profit increment and the investment amount to obtain the cost-effectiveness of each management sub-region;
[0155] Based on the aforementioned cost-effectiveness, the cost index for each management sub-region is obtained.
[0156] In some preferred embodiments, the cost index CI of the i-th management sub-region i Represented as:
[0157]
[0158] Among them, TP i TC represents the profit earned during the investment period of the i-th management sub-region.i The cost incurred during the investment period for the i-th management sub-region, TA i Let δ be the investment amount for the i-th management sub-region, and δ be the preset cost adjustment coefficient.
[0159] By calculating cost-benefit ratio coefficients, companies can identify potential cost savings and opportunities for improved efficiency. For example, by optimizing production processes, reducing procurement costs, and improving product quality, companies can enhance product competitiveness and market share without increasing or minimizing costs. This helps companies achieve sustained economic growth.
[0160] Further, preferably, when performing step S35, the resource comprehensive optimization index RC of the i-th management sub-region is... i Represented as:
[0161] RC i =RU i ×DE i ×TI i ×CI i ×θ (9);
[0162] Where θ is a preset resource comprehensive optimization adjustment coefficient.
[0163] In a preferred embodiment, after performing step S3, the cloud computing resource management method further includes:
[0164] Based on the comprehensive resource optimization index and the preset early warning threshold, it is determined whether there is any abnormal usage of cloud computing resources, and an early warning signal is sent when abnormal usage occurs.
[0165] Furthermore, preferably, when judging cloud computing resources, not only is a threshold comparison made for the overall resource optimization index, but also threshold comparisons are made for the processor speed index, disk efficiency index, performance index and cost index respectively, so as to achieve multi-dimensional early warning.
[0166] For example, if the processor speed index is less than a preset first threshold, it is determined that the enterprise's cloud computing resource utilization is abnormal. The abnormal result is sent to the operators and managers, and an early warning signal is issued. Otherwise, it indicates that the enterprise's cloud computing resource utilization is normal. If the disk efficiency index is less than a preset second threshold, it is determined that the enterprise's cloud computing disk efficiency is abnormal. The abnormal result is sent to the operators and managers, and an early warning signal is issued. Otherwise, it indicates that the enterprise's cloud computing disk efficiency is normal. If the performance index is less than a preset third threshold, it is determined that the enterprise's cloud computing performance is abnormal. The abnormal result is sent to the operators and managers, and an early warning signal is issued. Otherwise, it indicates that the enterprise's cloud computing performance is normal. If the cost-effectiveness index coefficient is less than a preset fourth threshold, it is determined that the enterprise's cloud computing cost-effectiveness is abnormal. The abnormal result is sent to the operators and managers, and an early warning signal is issued. Otherwise, it indicates that the enterprise's cloud computing cost-effectiveness is normal.
[0167] As a preferred implementation, step S4, allocating and managing resources among the management sub-regions according to the comprehensive resource optimization index, includes:
[0168] If the comprehensive resource optimization index of the managed sub-region is less than the preset optimization index threshold, the managed sub-region will be regarded as a sub-region to be optimized.
[0169] A resource optimization strategy is formulated that includes the sub-region to be optimized, and the resource optimization strategy is sent to each management sub-region to allocate and manage resources among the management sub-regions; the resource optimization strategy includes the addition, reduction, migration and load balancing of cloud computing resources.
[0170] For example, when the resource comprehensive optimization index RC of the i-th management sub-region i When the resource utilization rate of the i-th management sub-region is greater than the preset optimization index threshold Δθ, it indicates that the enterprise cloud computing resources of the i-th management sub-region are being utilized effectively, and data collection and analysis of each management sub-region should continue. i If the value is less than the preset optimization index threshold Δθ, it indicates that the enterprise cloud computing resource utilization effect of the i-th management sub-region is poor. In this case, an analysis report will be generated for the data of the i-th management sub-region and an early warning signal will be issued. At the same time, the i-th management sub-region will be regarded as a sub-region to be optimized.
[0171] It should be noted that those skilled in the art can locate the sub-region to be optimized after determining the comprehensive resource optimization index, so as to further generate resource optimization strategies. The specific resource optimization strategy generation method is not the focus of this invention embodiment, and those skilled in the art can adopt various resource optimization strategy generation methods based on existing technologies. The main idea of various cloud computing resource allocation methods is that when the load of a certain computing node is too high, the module will automatically migrate some tasks to other computing nodes to balance the load; when the resource utilization of a certain computing node is low, the module will appropriately reduce the resources allocated to it to avoid resource waste.
[0172] The cloud computing resource management method provided by this invention divides management sub-regions based on target main resources, laying the foundation for subsequent refined management. By collecting multi-dimensional computing power demand information parameters, the resource operation and demand status of each management sub-region can be fully grasped. This transforms the complex resource status of each sub-region into a quantitative resource comprehensive optimization index that can be directly used for decision-making, promoting the rational flow and allocation of resources among sub-regions. It can focus on the computing power characteristics of the target main resources and realize refined dynamic management of resources at the sub-region level.
[0173] This invention provides a cloud computing resource management system. Please refer to [link / reference]. Figure 2 The cloud computing resource management system includes a sub-region partitioning module 11, a parameter acquisition module 12, an index calculation module 13, and a resource optimization module 14, wherein:
[0174] Sub-region division module 11 is used to acquire target entity resources and divide the target management area where the target entity resources are located into several management sub-regions;
[0175] Parameter acquisition module 12 is used to collect computing power demand information parameters for each management sub-region; the computing power demand information parameters include processor operating parameters, disk operating parameters, throughput parameters and cost parameters;
[0176] The index calculation module 13 is used to calculate the comprehensive resource optimization index of each management sub-region based on the computing power demand information parameters.
[0177] The resource optimization module 14 is used to allocate and manage resources among the managed sub-regions based on the comprehensive resource optimization index.
[0178] In a preferred embodiment, the sub-region division module 11 includes:
[0179] The region determination unit is used to acquire target entity resources and determine the region where the target entity resources are located as the target management region.
[0180] The resource grouping unit is used to group the target main resource according to the logical attributes and operational requirements of the computing nodes in the target main resource;
[0181] The sub-region division unit is used to divide the target management area into several management sub-regions according to the grouping results of the target subject resources; each management sub-region includes at least one computing node.
[0182] In a preferred embodiment, the parameter acquisition module 12 includes:
[0183] The first parameter acquisition unit is used to acquire the temperature and humidity of each management sub-region when the processor is in operation, and to obtain the processor operating parameters;
[0184] The second parameter acquisition unit is used to collect the read speed, write speed and number of operations per second of the disk in each management sub-area to obtain the disk operating parameters;
[0185] The third parameter acquisition unit is used to collect the storage throughput and computing throughput of each management sub-region to obtain the throughput parameter.
[0186] The fourth parameter acquisition unit is used to collect the profit, cost, and investment amount of each management sub-region to obtain cost parameters;
[0187] The parameter aggregation unit is used to obtain computing power demand information parameters based on the processor operating parameters, the disk operating parameters, the throughput parameters, and the cost parameters.
[0188] In a preferred embodiment, the index calculation module 13 includes:
[0189] The processor rate index calculation unit is used to calculate the rate deviation of each management sub-region based on the processor operating parameters, and use the rate deviation as a standard normal distribution variable to obtain the processor rate index.
[0190] The disk efficiency index calculation unit is used to calculate the disk utilization rate of each management sub-region based on the disk operating parameters, and to obtain the disk efficiency index based on the non-linear growth relationship between the disk utilization rate and disk efficiency.
[0191] The performance index calculation unit is used to calculate the collaborative relationship between the throughput of each management sub-region and the time correction based on the throughput parameter, and obtain the performance index.
[0192] The cost index calculation unit is used to calculate the cost-effectiveness of each management sub-region based on the cost parameters, and obtain the cost index.
[0193] The resource comprehensive optimization index calculation unit is used to obtain the resource comprehensive optimization index of each management sub-region based on the processor speed index, the disk efficiency index, the performance index, and the cost index.
[0194] Further, preferably, the processor operating parameters include the temperature and humidity of each management sub-region where the processor is in operation; then the processor rate index calculation unit is specifically used for:
[0195] The total temperature of all management sub-regions is obtained based on the temperature of each management sub-region where the processor is in operation.
[0196] Based on the temperature, the total temperature, and the total processor running time, the difference between the temperature of each management sub-region and the total temperature is calculated to obtain the rate temperature deviation.
[0197] The average humidity of each management sub-region is obtained based on the humidity of the processors in operation within each management sub-region.
[0198] Based on the humidity and the average humidity, the difference between the humidity of each managed sub-area and the average humidity is calculated to obtain the rate humidity deviation.
[0199] The rate deviation is obtained by multiplying the rate-temperature deviation and the rate-humidity deviation.
[0200] The processor rate index is calculated by treating the rate deviation as a standard normal distribution variable.
[0201] Preferably, the disk operating parameters include the disk's read speed, write speed, and operations per second; the disk efficiency index calculation unit is specifically used for:
[0202] The sum of the read speed and write speed of each management sub-region is calculated to obtain the total disk transfer rate;
[0203] The quotient of the sum of the disk transfer rates and the number of operations per second is used as the disk utilization rate;
[0204] The disk utilization rate is incremented by 1 and then logarithmically calculated to obtain the disk efficiency index.
[0205] Preferably, the throughput parameters include storage throughput and computing throughput; the performance index calculation unit is specifically used for:
[0206] The average storage throughput is calculated based on the storage throughput and total time of each management sub-region;
[0207] The average computing throughput is calculated based on the computing throughput and total time of each management sub-region.
[0208] Based on the storage throughput, the average storage throughput, the computing throughput, and the average computing throughput, a relative synergistic relationship between computing throughput and storage throughput is obtained.
[0209] The performance index is derived based on the relative synergistic relationship between computation throughput and storage throughput.
[0210] Preferably, the cost parameters include profit, cost, and investment amount; then the cost index calculation unit is specifically used for:
[0211] Calculate the difference between the profit and the cost for each management sub-region to obtain the net profit increment;
[0212] Calculate the quotient of the net profit increment and the investment amount to obtain the cost-effectiveness of each management sub-region;
[0213] Based on the aforementioned cost-effectiveness, the cost index for each management sub-region is obtained.
[0214] In a preferred embodiment, the cloud computing resource management system further includes an early warning module, which responds to the index calculation module 13 and is used to:
[0215] Based on the comprehensive resource optimization index and the preset early warning threshold, it is determined whether there is any abnormal usage of cloud computing resources, and an early warning signal is sent when abnormal usage occurs.
[0216] In a preferred embodiment, the resource optimization module 14 is specifically used for:
[0217] If the comprehensive resource optimization index of the managed sub-region is less than the preset optimization index threshold, the managed sub-region will be regarded as a sub-region to be optimized.
[0218] A resource optimization strategy is formulated that includes the sub-region to be optimized, and the resource optimization strategy is sent to each management sub-region to allocate and manage resources among the management sub-regions; the resource optimization strategy includes the addition, reduction, migration and load balancing of cloud computing resources.
[0219] The cloud computing resource management system provided by this invention divides management sub-regions based on target main resources, laying the foundation for subsequent refined management. By collecting multi-dimensional computing power demand information parameters, it can comprehensively grasp the resource operation and demand of each management sub-region, transforming the complex resource status of each sub-region into a quantitative resource comprehensive optimization index that can be directly used for decision-making. This promotes the rational flow and allocation of resources among sub-regions, focuses on the computing power characteristics of the target main resources, and realizes refined dynamic management of resources at the sub-region level.
[0220] Please see Figure 3 , Figure 3 This is a structural block diagram of a cloud computing resource management device provided in an embodiment of the present invention. The cloud computing resource management includes a processor 31, a memory 32, and a computer program stored in the memory 32 and executable on the processor 31. When the processor 31 executes the computer program, it implements the steps in the above-described embodiments of the various cloud computing resource management methods, such as steps S1 to S4.
[0221] For example, the computer program can be divided into one or more modules / units, which are stored in the memory 32 and executed by the processor 31 to complete the present invention. The one or more modules / units can be a series of computer program instruction segments capable of performing specific functions, which describe the execution process of the computer program in the cloud computing resource management device.
[0222] The cloud computing resource management device may include, but is not limited to, a processor 31 and a memory 32. Those skilled in the art will understand that the schematic diagram is merely an example of a cloud computing resource management device and does not constitute a limitation on the device. It may include more or fewer components than illustrated, or combine certain components, or use different components. For example, the cloud computing resource management device may also include input / output devices, network access devices, buses, etc.
[0223] The processor 31 can be a Central Processing Unit (CPU), or other general-purpose processors, digital signal processors (DSPs), application-specific integrated circuits (ASICs), field-programmable gate arrays (FPGAs), or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc. The general-purpose processor can be a microprocessor or any conventional processor. The processor 31 is the control center of the cloud computing resource management device, connecting various parts of the entire cloud computing resource management device through various interfaces and lines.
[0224] The memory 32 can be used to store the computer programs and / or modules. The processor 31 implements various functions of the cloud computing resource management device by running or executing the computer programs and / or modules stored in the memory 32 and calling the data stored in the memory 32. The memory 32 may mainly include a program storage area and a data storage area. The program storage area may store the operating system, at least one application program required for a function (such as sound playback function, image playback function, etc.), etc.; the data storage area may store data created according to the use of the mobile phone (such as audio data, phonebook, etc.). In addition, the memory 32 may include high-speed random access memory, and may also include non-volatile memory, such as hard disk, memory, plug-in hard disk, smart media card (SMC), secure digital card (SD) card, flash card, at least one disk storage device, flash memory device, or other volatile solid-state storage device.
[0225] Wherein, if the modules / units integrated in the cloud computing resource management device are implemented as software functional units and sold or used as independent products, they can be stored in a computer-readable storage medium. Based on this understanding, all or part of the processes in the methods of the above embodiments of the present invention can also be implemented by a computer program instructing related hardware. The computer program can be stored in a computer-readable storage medium, and when the computer program is executed by the processor 31, it can implement the steps of the various method embodiments described above. Wherein, the computer program includes computer program code, which can be in the form of source code, object code, executable file, or some intermediate form, etc. The computer-readable medium can include: any entity or device capable of carrying the computer program code, recording medium, USB flash drive, portable hard drive, magnetic disk, optical disk, computer memory, read-only memory (ROM), random access memory (RAM), electrical carrier signal, telecommunication signal, and software distribution medium, etc.
[0226] The above description represents the preferred embodiments of the present invention. It should be noted that those skilled in the art can make various improvements and modifications without departing from the principles of the present invention, and these improvements and modifications are also considered to be within the scope of protection of the present invention.
Claims
1. A cloud computing resource management method, characterized in that, include: Acquire the target entity resources, and divide the target management area where the target entity resources are located into several management sub-areas; Collect computing power demand information parameters for each management sub-region; the computing power demand information parameters include processor operating parameters, disk operating parameters, throughput parameters, and cost parameters; Based on the computing power demand information parameters, calculate the comprehensive resource optimization index for each management sub-region; Based on the aforementioned comprehensive resource optimization index, resources are allocated and managed among the sub-regions.
2. The cloud computing resource management method as described in claim 1, characterized in that, The acquisition of target entity resources involves dividing the target management area where the target entity resources are located into several management sub-areas, including: Acquire the target entity's resources and determine the area where the target entity's resources are located as the target management area; Based on the logical attributes and operational requirements of the computing nodes in the target entity resources, the target entity resources are grouped. Based on the grouping results of the target entity resources, the target management area is divided into several management sub-areas; each management sub-area includes at least one computing node.
3. The cloud computing resource management method as described in claim 1, characterized in that, The parameters for collecting computing power demand information for each management sub-region include: The temperature and humidity of each management sub-area where the processor is running are collected to obtain the processor's operating parameters; Collect the disk read speed, write speed, and operations per second of each managed sub-region to obtain disk operating parameters; Collect the storage throughput and computing throughput of each management sub-region to obtain throughput parameters; Collect the profit, cost, and investment amount of each management sub-region to obtain cost parameters; The computing power requirement information parameters are obtained based on the processor operating parameters, the disk operating parameters, the throughput parameters, and the cost parameters.
4. The cloud computing resource management method as described in claim 1, characterized in that, The step of calculating the comprehensive resource optimization index for each management sub-region based on the computing power demand information parameters includes: Based on the processor operating parameters, the rate deviation of each management sub-region is calculated, and the rate deviation is used as a standard normal distribution variable to obtain the processor rate index. Based on the disk operating parameters, the disk utilization rate of each management sub-region is calculated, and based on the non-linear growth relationship between the disk utilization rate and disk efficiency, the disk efficiency index is obtained. Based on the throughput parameters, the collaborative relationship between the throughput of each management sub-region and time correction is calculated to obtain the performance index; Based on the aforementioned cost parameters, the cost-effectiveness of each management sub-region is calculated to obtain a cost index; Based on the processor speed index, the disk efficiency index, the performance index, and the cost index, the comprehensive resource optimization index for each management sub-region is obtained.
5. A cloud computing resource management method as described in claim 4, characterized in that, The processor operating parameters include the temperature and humidity of each management sub-region where the processor is running; then, based on the processor operating parameters, the rate deviation of each management sub-region is calculated, and the rate deviation is used as a standard normal distribution variable to obtain the processor rate index, including: The total temperature of all management sub-regions is obtained based on the temperature of each management sub-region where the processor is in operation. Based on the temperature, the total temperature, and the total processor running time, the difference between the temperature of each management sub-region and the total temperature is calculated to obtain the rate temperature deviation. The average humidity of each management sub-region is obtained based on the humidity of the processors in operation within each management sub-region. Based on the humidity and the average humidity, the difference between the humidity of each managed sub-area and the average humidity is calculated to obtain the rate humidity deviation. The rate deviation is obtained by multiplying the rate-temperature deviation and the rate-humidity deviation. The processor rate index is calculated by treating the rate deviation as a standard normal distribution variable.
6. The cloud computing resource management method as described in claim 4, characterized in that, The disk operating parameters include disk read speed, write speed, and operations per second; then, based on the disk operating parameters, the disk utilization of each management sub-region is calculated, and based on the non-linear growth relationship between disk utilization and disk efficiency, a disk efficiency index is obtained, including: The sum of the read speed and write speed of each management sub-region is calculated to obtain the total disk transfer rate; The quotient of the sum of the disk transfer rates and the number of operations per second is used as the disk utilization rate; The disk utilization rate is incremented by 1 and then logarithmically calculated to obtain the disk efficiency index.
7. A cloud computing resource management method as described in claim 4, characterized in that, The throughput parameters include storage throughput and computing throughput; then, based on the throughput parameters, the collaborative relationship between the throughput of each management sub-region and time-adjusted is calculated to obtain a performance index, including: The average storage throughput is calculated based on the storage throughput and total time of each management sub-region; The average computing throughput is calculated based on the computing throughput and total time of each management sub-region. Based on the storage throughput, the average storage throughput, the computing throughput, and the average computing throughput, a relative synergistic relationship between computing throughput and storage throughput is obtained. The performance index is derived based on the relative synergistic relationship between computation throughput and storage throughput.
8. The cloud computing resource management method as described in claim 4, characterized in that, The cost parameters include profit, cost, and investment amount; therefore, the cost-effectiveness of each management sub-region is calculated based on the cost parameters to obtain a cost index, including: Calculate the difference between the profit and the cost for each management sub-region to obtain the net profit increment; Calculate the quotient of the net profit increment and the investment amount to obtain the cost-effectiveness of each management sub-region; Based on the aforementioned cost-effectiveness, the cost index for each management sub-region is obtained.
9. A cloud computing resource management method as described in claim 1, characterized in that, After calculating the comprehensive resource optimization index for each management sub-region based on the computing power demand information parameters, the cloud computing resource management method further includes: Based on the comprehensive resource optimization index and the preset early warning threshold, it is determined whether there is any abnormal usage of cloud computing resources, and an early warning signal is sent when abnormal usage occurs.
10. A cloud computing resource management method as described in claim 1, characterized in that, The allocation and management of resources among the management sub-regions based on the comprehensive resource optimization index includes: If the comprehensive resource optimization index of the managed sub-region is less than the preset optimization index threshold, the managed sub-region will be regarded as a sub-region to be optimized. A resource optimization strategy is formulated, including the sub-regions to be optimized, and the resource optimization strategy is sent to each management sub-region to allocate and manage resources among the management sub-regions; the resource optimization strategy includes the addition, reduction, migration, and load balancing of cloud computing resources.
11. A cloud computing resource management system, characterized in that, include: The sub-region division module is used to acquire target entity resources and divide the target management area where the target entity resources are located into several management sub-regions. The parameter acquisition module is used to collect computing power demand information parameters for each management sub-region; the computing power demand information parameters include processor operating parameters, disk operating parameters, throughput parameters, and cost parameters; The index calculation module is used to calculate the comprehensive resource optimization index of each management sub-region based on the computing power demand information parameters. The resource optimization module is used to allocate and manage resources among the managed sub-regions based on the comprehensive resource optimization index.
12. A cloud computing resource management device, characterized in that, The system includes a processor, a memory, and a computer program stored in the memory and configured to be executed by the processor, wherein the processor, when executing the computer program, implements the cloud computing resource management method as described in any one of claims 1 to 10.
13. A computer-readable storage medium, characterized in that, The computer-readable storage medium includes a stored computer program, wherein, when the computer program is executed, it controls the device where the computer-readable storage medium is located to perform the cloud computing resource management method as described in any one of claims 1 to 10.
14. A computer program product, characterized in that, The computer program product includes a computer program or computer instructions, which, when executed by a processor, perform the cloud computing resource management method as described in any one of claims 1 to 10.