Data center-oriented resource optimization management method and system
Through the resource optimization management system for data centers, using resource analysis, optimization judgment, simulation modules and other means, the problem of difficult data center resource management is solved, and the rapid and reasonable optimization of resources and the improvement of system performance is achieved.
Patent Information
- Application Number
- CN202411771855.4
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2024-12-04
- Publication Date
- 2025-03-25
- Estimated Expiration
- Not applicable · inactive patent
AI Technical Summary
The existing data center resource management methods are difficult to achieve refined management and efficient utilization, resulting in waste of resources and degradation of performance, and lack of effective simulation verification methods, which makes optimization results difficult to predict.
Provide a resource optimization management system for data centers, including resource analysis module, resource optimization judgment module, resource optimization simulation module and resource optimization management module. By regularly updating core resource analysis values, dividing good table resources, poor table resources and general table resources, building resource optimization simulation models, generating optimization solutions and performing simulation verification.
It realizes rapid and reasonable optimization of data center resources, improves resource utilization and system performance, reduces resource waste and conflicts, and improves the efficiency and predictability of resource optimization management.
Smart Images
Figure CN119690668A_ABST
Abstract
Description
Technical Field
[0001] The present invention relates to the technical field of data centers, and more specifically, to a resource optimization management method and system for data centers. Background Art
[0002] With the rapid development of information technology, data centers, as core facilities for data storage, processing and transmission, are becoming increasingly important. Data centers not only bear the task of storing and managing massive amounts of data, but are also directly related to business continuity and stability. However, with the explosive growth of data volume and the diversification of business needs, data centers are facing unprecedented challenges in resource management and optimization.
[0003] Traditional data center resource management often relies on manual experience and simple rule scheduling, which makes it difficult to achieve refined management and efficient use of resources. First, unreasonable resource allocation causes some resources to be idle for a long time, resulting in resource waste; second, during high concurrency or business peak periods, resource shortages may lead to service performance degradation or even interruption, seriously affecting user experience and business benefits; third, optimizing resources is likely to affect the rest of the resources in the data center, causing abnormalities in resources that do not need to be optimized.
[0004] In order to meet these challenges, the industry has begun to explore data center resource optimization management methods based on data analysis and intelligent algorithms. However, most existing methods only stay at the level of simple monitoring and statistics of resource usage, lacking the ability to deeply analyze resource characteristics and accurately predict them. In addition, in the process of formulating and implementing resource optimization plans, there is often a lack of effective simulation verification methods, which makes it difficult to predict the optimization effect and may even cause new resource conflicts and performance problems. Summary of the invention
[0005] In view of the deficiencies in the prior art, the present invention aims to provide a resource optimization management method and system for a data center.
[0006] To achieve the above object, the present invention provides the following technical solutions:
[0007] Resource optimization management system for data centers, including resource analysis module, resource optimization determination module, resource optimization simulation module, and resource optimization management module;
[0008] The resource analysis module is used to regularly update the core resource analysis value of each core resource of the data center, and divide the core resources of the data center into good table resources, poor table resources and general table resources;
[0009] The resource optimization determination module obtains a resource optimization management value of the data center according to the best meter resources, the poor meter resources, and the average meter resources, and determines whether to perform resource optimization management on the data center based on a comparison result between the resource optimization management value and the resource optimization management threshold;
[0010] After determining to perform resource optimization management on the data center, the resource optimization simulation module obtains the actual data set of core resource indicators and the target data set of core resource indicators of each core resource, and builds a resource optimization simulation model of the data center based on the actual data set of core resource indicators and the target data set of core resource indicators of each core resource;
[0011] The resource optimization management module controls the resource optimization simulation model of the data center to perform simulated resource optimization, and gradually generates multiple resource optimization plans. Each time a resource optimization plan is generated, a resource optimization target value of the resource optimization plan is obtained. Based on the comparison result of the resource optimization target value and the resource optimization target threshold, it is determined whether the resource optimization plan is a resource optimization target plan, and the resource optimization target plan is used to perform resource optimization management on the data center.
[0012] Furthermore, the core resources of the data center are divided into good meter resources, poor meter resources and general meter resources, specifically: set the core resource analysis upper value and the core resource analysis lower value, when the core resource analysis value of the core resource is greater than or equal to the core resource analysis upper value, mark the core resource as a good meter resource, when the core resource analysis value of the core resource is less than or equal to the core resource analysis lower value, mark the core resource as a poor meter resource, when the core resource analysis value of the core resource is between the core resource analysis upper value and the core resource analysis lower value, mark the core resource as a general meter resource, obtain the stage performance value of the general meter resource, set the stage performance limit value, when the stage performance value of the general meter resource is less than the stage performance limit value, mark the general meter resource as a poor meter resource.
[0013] Furthermore, the stage performance value of the general table resources is obtained by the following method: obtain a core resource analysis values updated by the general table resources at the current time of the system, sort all the core resource analysis values in the order of update time, sum the two adjacent core resource analysis values after sorting to obtain the core resource head and tail sum value, sum the head and tail sum values of all core resources and take the average value to obtain the core resource head and tail sum mean value fgm, calculate the difference between the two adjacent core resource analysis values after sorting and take the absolute value to obtain the core resource turbulence value, sum all the core resource turbulence values and take the average value to obtain the core resource turbulence mean value skz, and use the formula The stage performance value Zmg of the general resource is obtained, where ha is the head and tail coefficient of the core resource and hb is the turbulence coefficient of the core resource.
[0014] Furthermore, the core resource analysis value of the core resource is updated in the following manner: collecting corresponding data of each type of resource indicators of the core resource within a period of t, integrating and packaging the corresponding data of each type of resource indicators into an actual data set of core resource indicators, obtaining the core resource analysis model of the core resource, using the actual data set of core resource indicators as input data of the core resource analysis model, and the core resource analysis model outputs the core resource analysis value of the core resource.
[0015] Furthermore, the resource optimization management value of the data center is obtained according to the good table resources, poor table resources and general table resources. Specifically, the total number of poor table resources is marked as HSp, the core resource analysis value of the poor table resources is marked as STe, e = 1, 2, ..., E, E is the total number of poor table resources, the core resource analysis coefficient is set to HAw, and the formula is used Get the resource optimization management value EAtd of the data center.
[0016] Furthermore, based on the comparison result between the resource optimization management value and the resource optimization management threshold, it is determined whether to perform resource optimization management on the data center. Specifically, a resource optimization management threshold is set, and when the resource optimization management value of the data center is greater than or equal to the resource optimization management threshold, it is determined to perform resource optimization management on the data center.
[0017] Furthermore, the resource optimization target value of the resource optimization scheme is obtained in the following way: when the resource optimization simulation model uses the resource optimization scheme to simulate resource optimization of the data center, the resource optimization management value of the data center is obtained every t time. When the resource optimization scheme completes the simulation resource optimization of the data center, the total number of resource optimization management values is marked as FCas, and all resource optimization management values are sorted in the order of acquisition time. The ratio of the previous adjacent resource optimization management value and the next adjacent resource optimization management value after sorting is calculated to obtain the resource optimization neighbor ratio value, and the resource optimization neighbor ratio threshold is set. When the resource optimization neighbor ratio value is greater than or equal to the resource optimization neighbor ratio threshold, the number of excellent resources is increased by one. When the resource optimization neighbor ratio value is less than the resource optimization neighbor ratio threshold, the number of excellent resources is increased by one, and the number of excellent resources is marked as STge, and the number of excellent resources is marked as ASkm. All resource optimization management values are summed and averaged to obtain the resource optimization management mean MPzw. The formula is used The reasonable resource optimization value YHHn of the resource optimization scheme is obtained, where ga is the resource optimization quantity coefficient, gb is the resource optimal quantity coefficient, gc is the resource excellent quantity coefficient, and gd is the optimization management mean coefficient. The reasonable resource optimization values of all historical resource optimization schemes are summed and averaged to obtain the reasonable historical optimization value GSXz. Obtain the resource optimization target value BZET.
[0018] Furthermore, based on the comparison result between the resource optimization target value and the resource optimization target threshold, determine whether the resource optimization plan is a resource optimization target plan, specifically: set the resource optimization target threshold, when the resource optimization target value of the resource optimization plan is greater than or equal to the resource optimization target threshold, stop generating the resource optimization plan, and mark the resource optimization plan as the resource optimization target plan; when the resource optimization target value of the resource optimization plan is less than the resource optimization target threshold, mark the resource optimization plan as a resource optimization historical plan, and continue to generate the next resource optimization plan.
[0019] Furthermore, the resource optimization management method for data center includes the following steps:
[0020] Step 1: Regularly update the core resource analysis values of each core resource in the data center, and divide the core resources of the data center into good table resources, poor table resources, and general table resources;
[0021] Step 2: Obtain a resource optimization management value of the data center according to the best meter resources, the poor meter resources, and the average meter resources, and determine whether to perform resource optimization management on the data center based on a comparison result between the resource optimization management value and the resource optimization management threshold;
[0022] Step 3: After determining to perform resource optimization management on the data center, obtain the actual data set of core resource indicators and the target data set of core resource indicators of each core resource, and build a resource optimization simulation model of the data center based on the actual data set of core resource indicators and the target data set of core resource indicators of each core resource;
[0023] Step 4: Control the resource optimization simulation model of the data center to perform simulation resource optimization and gradually generate multiple resource optimization solutions;
[0024] Step 5: Each time a resource optimization plan is generated, the resource optimization target value of the resource optimization plan is obtained. Based on the comparison result between the resource optimization target value and the resource optimization target threshold, it is determined whether the resource optimization plan is a resource optimization target plan, and the resource optimization target plan is used to perform resource optimization management on the data center.
[0025] Compared with the prior art, the present invention has the following beneficial effects:
[0026] 1. The method of the present invention uses good meter resources, poor meter resources and general meter resources to three-dimensionally analyze whether it is necessary to optimize the resource management of the data center, and quickly determine the resource optimization plan suitable for the data center;
[0027] 2. Set up a resource analysis module and a resource optimization judgment module to conduct an in-depth analysis of various core resources in the data center, and divide the core resources into good table resources, poor table resources and general table resources, and conduct a secondary analysis of the general table resources, conduct an in-depth understanding and analysis of the stage performance of the general table resources, and deeply analyze whether the general table resources are poor table resources. Set up a resource optimization simulation module and a resource optimization management module. When it is necessary to optimize the resources of the data center, simulate and optimize the resources by building a resource optimization simulation model, and then generate a resource optimization plan. Comprehensively compare and analyze the generated resource optimization plans, reasonably control the number of iterations of the resource optimization simulation model, quickly and reasonably complete the resource optimization work of the data center, and improve overall work efficiency. BRIEF DESCRIPTION OF THE DRAWINGS
[0028] Figure 1 A method flow chart of a resource optimization management method for a data center;
[0029] Figure 2 This is the principle block diagram of the resource optimization management module;
[0030] Figure 3 This is a system module diagram of a resource optimization management system for a data center. DETAILED DESCRIPTION
[0031] Example 1: Reference Figure 1 , a resource optimization management method for a data center includes the following steps:
[0032] Step 1: Regularly update the core resource analysis values of each core resource in the data center, and divide the core resources of the data center into good table resources, poor table resources, and general table resources.
[0033] Step 2: Obtain the resource optimization management value of the data center according to the best meter resources, the poor meter resources, and the average meter resources, and determine whether to perform resource optimization management on the data center based on the comparison result between the resource optimization management value and the resource optimization management threshold.
[0034] Step 3: After deciding to optimize the resources of the data center, obtain the actual data set of core resource indicators and the target data set of core resource indicators of each core resource, and build a resource optimization simulation model of the data center based on the actual data set of core resource indicators and the target data set of core resource indicators of each core resource.
[0035] Step 4: Control the resource optimization simulation model of the data center to perform simulation resource optimization and gradually generate multiple resource optimization solutions.
[0036] Step 5: Each time a resource optimization plan is generated, the resource optimization target value of the resource optimization plan is obtained. Based on the comparison result between the resource optimization target value and the resource optimization target threshold, it is determined whether the resource optimization plan is a resource optimization target plan, and the resource optimization target plan is used to perform resource optimization management on the data center.
[0037] The above method uses the best meter resources, poor meter resources and average meter resources to three-dimensionally analyze whether it is necessary to optimize the resource management of the data center, and quickly determine the resource optimization plan suitable for the data center.
[0038] Example 2: Reference Figure 2-Figure 3 , a resource optimization management system for data centers, including a resource analysis module, a resource optimization determination module, a resource optimization simulation module, and a resource optimization management module.
[0039] Resource analysis module: update the core resource analysis value of each core resource of the data center every t time (the core resources of the data center include but are not limited to CPU resources, memory resources, storage resources, network resources, and energy resources), set the core resource analysis upper value and the core resource analysis lower value (the core resource analysis upper value is greater than the core resource analysis lower value, and the core resource analysis upper value and the core resource analysis lower value are both set thresholds), when the core resource analysis value of the core resource is greater than or equal to the core resource analysis upper value, mark the core resource as a good meter resource, when the core resource analysis value of the core resource is less than or equal to the core resource analysis lower value, mark the core resource as a poor meter resource, when the core resource analysis value of the core resource is between the core resource analysis upper value and the core resource analysis lower value, mark the core resource as a general meter resource, obtain the stage performance value of the general meter resource, set the stage performance limit (the stage performance limit is the set threshold), when the stage performance value of the general meter resource is greater than or equal to the stage performance limit, no processing is performed, when the stage performance value of the general meter resource is less than the stage performance limit, mark the general meter resource as a poor meter resource.
[0040] The core resource analysis value of the core resource is updated in the following way: collect the corresponding data of various types of resource indicators of the core resource within a time period of t (different core resources correspond to different types of resource indicators. Taking CPU resources as an example, the resource indicators of CPU resources include but are not limited to CPU utilization, CPU usage rate, and CPU load. Taking memory resources as an example, the resource indicators of memory resources include but are not limited to memory utilization, memory latency, and memory access speed), integrate and package the corresponding data of various types of resource indicators into an actual data set of core resource indicators, obtain the core resource analysis model of the core resource, use the actual data set of core resource indicators as the input data of the core resource analysis model, and the core resource analysis model outputs the core resource analysis value of the core resource.
[0041] The corresponding data of various types of resource indicators are integrated and packaged into the actual data set of core resource indicators. Specifically, the corresponding data of various types of resource indicators are cleaned, including removing duplicate data, filling missing values, correcting erroneous data, etc. The cleaned data are formatted and processed, and the cleaned and converted data are merged together to form the actual data set of core resource indicators.
[0042] Different core resources correspond to different core resource analysis models. For example, CPU resources and memory resources correspond to a core resource analysis model respectively. The construction methods of different core resource analysis models are similar, except that the corresponding data in the construction process is replaced. This embodiment takes CPU resources as an example to introduce the construction process of the core resource analysis model: collect actual data sets of core resource indicators of multiple CPU resources (if it is a core resource analysis model of memory resources, it is necessary to collect actual data sets of core resource indicators of multiple memory resources), build a deep learning model, use the actual data sets of core resource indicators of CPU resources as training data of the deep learning model, assign a core resource analysis value to each training data, and the value range of the core resource analysis value is (5-10). The larger the value of the core resource analysis value, the worse the resource performance of the CPU resource (if it is a core resource analysis model of memory resources, the larger the value of the core resource analysis value, the worse the resource performance of the memory resource), and the smaller the value of the core resource analysis value, the better the resource performance of the CPU resource. The training data is divided into a 70% training set and a 30% validation set, and the training set and the validation set are trained. After the training is completed, the core resource analysis model of the CPU resource is obtained.
[0043] The stage performance value of general table resources is obtained by the following method: obtain a core resource analysis values updated by general table resources at the current time of the system, sort all core resource analysis values in the order of update time, sum the two adjacent core resource analysis values after sorting to obtain the core resource head and tail sum value, sum the head and tail sum values of all core resources and take the average value to obtain the core resource head and tail sum mean fgm, calculate the difference between the two adjacent core resource analysis values after sorting and take the absolute value to obtain the core resource turbulence value, sum all core resource turbulence values and take the average value to obtain the core resource turbulence mean skz, and use the formula The stage performance value Zmg of the general resource is obtained, where ha is the head and tail coefficient of the core resource, hb is the turbulence coefficient of the core resource, the value of ha is 0.28, and the value of hb is 0.24.
[0044] Resource optimization determination module: mark the total number of difference table resources as HSp, mark the core resource analysis value of the difference table resources as STe, e=1, 2, ..., E, E is the total number of difference table resources, set the core resource analysis coefficient as HAw, w=1, 2, 3, ...w; HA1<HA2<HA3<...<HAw, each core resource analysis coefficient corresponds to a range of core resource analysis values, the range of core resource analysis values includes (0, ST1], (ST1, ST2], ..., (STE-1, STE], when STE∈(0, ST1], the core resource analysis coefficient is HA1, using the formula Get the resource optimization management value EAtd of the data center, set the resource optimization management threshold (the resource optimization management threshold is the set threshold), when the resource optimization management value of the data center is greater than or equal to the resource optimization management threshold, determine to perform resource optimization management on the data center, when the resource optimization management value of the data center is less than the resource optimization management threshold, no processing is performed.
[0045] By setting up a resource analysis module and a resource optimization judgment module, we can conduct an in-depth analysis of various core resources in the data center, and divide the core resources into good meter resources, poor meter resources and general meter resources. We can also conduct a secondary analysis of the general meter resources, conduct an in-depth understanding and analysis of the stage performance of the general meter resources, and deeply analyze whether the general meter resources are poor meter resources. Through a three-dimensional analysis of good meter resources, poor meter resources and general meter resources, we can determine whether it is necessary to optimize the resource management of the data center.
[0046] Resource optimization simulation module: After determining to optimize the resources of the data center, obtain the actual data set of core resource indicators and the target data set of core resource indicators for each core resource (taking CPU resources as an example, the actual data set of core resource indicators is the integrated packaged data set of various types of resource indicator data generated by CPU resources within a period of t, and the target data set of core resource indicators is the data set preset by the system for various types of resource indicator data of CPU resources), and build a resource optimization simulation model of the data center based on the actual data set of core resource indicators and the target data set of core resource indicators of each core resource.
[0047] A resource optimization simulation model of the data center is constructed based on the actual data sets of core resource indicators and the target data sets of core resource indicators of each core resource. Specifically, visual simulation tools such as Data Center Design Pro and MATLAB Simu link are selected to establish the framework of the simulation model according to the processing and distribution of core resources in the data center. The parameters and constraints of the simulation model are set according to the actual data sets of core resource indicators and the target data sets of core resource indicators to construct the resource optimization simulation model of the data center.
[0048] Resource optimization management module: The resource optimization simulation model of the data center performs simulated resource optimization and gradually generates multiple resource optimization plans. Each time a resource optimization plan is generated, the resource optimization target value of the resource optimization plan is obtained, and the resource optimization target threshold is set (the resource optimization target threshold is the set threshold). When the resource optimization target value of the resource optimization plan is greater than or equal to the resource optimization target threshold, stop generating resource optimization plans and mark the resource optimization plan as a resource optimization target plan. Use the resource optimization target plan to perform resource optimization management on the data center. When the resource optimization target value of the resource optimization plan is less than the resource optimization target threshold, mark the resource optimization plan as a resource optimization history plan and continue to generate the next resource optimization plan.
[0049] The resource optimization target value of the resource optimization scheme is obtained in the following way: when the resource optimization simulation model uses the resource optimization scheme to simulate resource optimization of the data center, the resource optimization management value of the data center is obtained every t time. When the resource optimization scheme completes the simulation resource optimization of the data center, the total number of resource optimization management values is marked as FCas, and all resource optimization management values are sorted in the order of acquisition time. The ratio of the previous adjacent resource optimization management value and the next adjacent resource optimization management value after sorting is calculated to obtain the resource optimization neighbor ratio value, and the resource optimization neighbor ratio threshold is set (the resource optimization neighbor ratio threshold is the set threshold). When the resource optimization neighbor ratio value is greater than or equal to the resource optimization neighbor ratio threshold, the number of excellent resources is increased by one. When the resource optimization neighbor ratio value is less than the resource optimization neighbor ratio threshold, the number of excellent resources is increased by one, the number of excellent resources is marked as STge, and the number of excellent resources is marked as ASkm. All resource optimization management values are summed and averaged to obtain the resource optimization management mean MPzw. The formula is used The resource optimization reasonable value YHHn of the resource optimization scheme is obtained, where ga is the resource optimization quantity coefficient, gb is the resource optimal quantity coefficient, gc is the resource excellent quantity coefficient, and gd is the optimization management mean coefficient. The value of ga is 0.52, the value of gb is 1.31, the value of gc is 0.68, and the value of gd is 0.63. The resource optimization reasonable values of all resource optimization historical schemes are summed and averaged to obtain the historical optimization reasonable value GSXz. Using the formula Obtain the resource optimization target value BZET.
[0050] Set up a resource optimization simulation module and a resource optimization management module. When it is necessary to optimize the resources of the data center, the resources are simulated and optimized by building a resource optimization simulation model, and then a resource optimization plan is generated. A comprehensive comparison and analysis of the generated resource optimization plans can be performed to quickly determine the resource optimization plan suitable for the data center, reasonably control the number of iterations of the resource optimization simulation model, quickly and reasonably complete the resource optimization work of the data center, and improve overall work efficiency.
[0051] The above formulas are all dimensionless and numerical calculations are performed, and the preset parameters in the formulas are set by technicians in this field according to actual conditions.
[0052] The above embodiments can be implemented in whole or in part by software, hardware, firmware or any other combination. When implemented by software, the above embodiments can be implemented in whole or in part in the form of a computer program product. The computer program product includes one or more computer instructions or computer programs. When the computer instructions or computer programs are loaded or executed on a computer, the process or function described in the embodiment of the present application is generated in whole or in part. The computer can be a general-purpose computer, a special-purpose computer, a computer network, or other programmable device. The computer instructions can be stored in a computer-readable storage medium, or transmitted from one computer-readable storage medium to another computer-readable storage medium. For example, the computer instructions can be transmitted from one website site, computer, server or data center to another website site, computer, server or data center by wired or wireless (e.g., infrared, wireless, microwave, etc.). The computer-readable storage medium can be any available medium that a computer can access or a data storage device such as a server or data center that contains one or more available media sets. The available medium can be a magnetic medium (e.g., a floppy disk, a hard disk, a tape), an optical medium (e.g., a DVD), or a semiconductor medium. The semiconductor medium can be a solid-state hard disk.
[0053] It should be understood that in the various embodiments of the present application, the size of the serial numbers of the above-mentioned processes does not mean the order of execution. The execution order of each process should be determined by its function and internal logic, and should not constitute any limitation on the implementation process of the embodiments of the present application.
[0054] Those of ordinary skill in the art will appreciate that the units and algorithm steps of each example described in conjunction with the embodiments disclosed herein can be implemented in electronic hardware, or a combination of computer software and electronic hardware. Whether these functions are performed in hardware or software depends on the specific application and design constraints of the technical solution. Professional and technical personnel can use different methods to implement the described functions for each specific application, but such implementation should not be considered to be beyond the scope of this application.
[0055] Those skilled in the art can clearly understand that, for the convenience and brevity of description, the specific working processes of the systems, devices and units described above can refer to the corresponding processes in the aforementioned method embodiments and will not be repeated here.
[0056] In the several embodiments provided in the present application, it should be understood that the disclosed systems, devices and methods can be implemented in other ways. For example, the device embodiments described above are only schematic. For example, the division of the units is only a logical function division. There may be other division methods in actual implementation, such as multiple units or components can be combined or integrated into another system, or some features can be ignored or not executed. Another point is that the mutual coupling or direct coupling or communication connection shown or discussed can be through some interfaces, indirect coupling or communication connection of devices or units, which can be electrical, mechanical or other forms.
[0057] If the functions are implemented in the form of software functional units and sold or used as independent products, they can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present application can be essentially or partly embodied in the form of a software product that contributes to the prior art. The computer software product is stored in a storage medium and includes several instructions for a computer device (which can be a personal computer, a server, or a network device, etc.) to perform all or part of the steps of the methods described in the various embodiments of the present application. The aforementioned storage media include: various media that can store program codes, such as USB flash drives, mobile hard disks, read-only memories (ROM), random access memories (RAM), magnetic disks or optical disks.
[0058] The above is only a specific implementation of the present application, but the protection scope of the present application is not limited thereto. Any person skilled in the art who is familiar with the present technical field can easily think of changes or substitutions within the technical scope disclosed in the present application, which should be included in the protection scope of the present application. Therefore, the protection scope of the present application should be based on the protection scope of the claims.
Claims
1. A resource optimization management system for data centers, characterized in that: It includes resource analysis module, resource optimization determination module, resource optimization simulation module and resource optimization management module; The resource analysis module is used to regularly update the core resource analysis value of each core resource of the data center, and divide the core resources of the data center into good table resources, poor table resources and general table resources; The resource optimization determination module obtains a resource optimization management value of the data center according to the best meter resources, the poor meter resources, and the average meter resources, and determines whether to perform resource optimization management on the data center based on a comparison result between the resource optimization management value and the resource optimization management threshold; After determining to perform resource optimization management on the data center, the resource optimization simulation module obtains the actual data set of core resource indicators and the target data set of core resource indicators of each core resource, and builds a resource optimization simulation model of the data center based on the actual data set of core resource indicators and the target data set of core resource indicators of each core resource; The resource optimization management module controls the resource optimization simulation model of the data center to perform simulated resource optimization, and gradually generates multiple resource optimization plans. Each time a resource optimization plan is generated, a resource optimization target value of the resource optimization plan is obtained. Based on the comparison result of the resource optimization target value and the resource optimization target threshold, it is determined whether the resource optimization plan is a resource optimization target plan, and the resource optimization target plan is used to perform resource optimization management on the data center.
2. The resource optimization management system for data centers according to claim 1, characterized in that: The core resources of the data center are divided into good meter resources, poor meter resources and general meter resources. Specifically, the core resource analysis upper value and the core resource analysis lower value are set. When the core resource analysis value of the core resource is greater than or equal to the core resource analysis upper value, the core resource is marked as a good meter resource. When the core resource analysis value of the core resource is less than or equal to the core resource analysis lower value, the core resource is marked as a poor meter resource. When the core resource analysis value of the core resource is between the core resource analysis upper value and the core resource analysis lower value, the core resource is marked as a general meter resource. The stage performance value of the general meter resource is obtained, and the stage performance limit is set. When the stage performance value of the general meter resource is less than the stage performance limit, the general meter resource is marked as a poor meter resource.
3. The resource optimization management system for data centers according to claim 2, characterized in that: The stage performance value of general table resources is obtained by the following method: obtain a core resource analysis values updated by general table resources at the current time of the system, sort all core resource analysis values in the order of update time, sum the two adjacent core resource analysis values after sorting to obtain the core resource head and tail sum value, sum the head and tail sum values of all core resources and take the average value to obtain the core resource head and tail sum mean fgm, calculate the difference between the two adjacent core resource analysis values after sorting and take the absolute value to obtain the core resource turbulence value, sum all core resource turbulence values and take the average value to obtain the core resource turbulence mean skz, and use the formula The stage performance value Zmg of the general resource is obtained, where ha is the head and tail coefficient of the core resource and hb is the turbulence coefficient of the core resource.
4. The resource optimization management system for data centers according to claim 3, characterized in that: The core resource analysis value of the core resource is updated in the following way: collect the corresponding data of each type of resource indicators of the core resource within a period of t, integrate and package the corresponding data of each type of resource indicators into an actual data set of core resource indicators, obtain the core resource analysis model of the core resource, use the actual data set of core resource indicators as input data of the core resource analysis model, and the core resource analysis model outputs the core resource analysis value of the core resource.
5. The resource optimization management system for data centers according to claim 1, characterized in that: The resource optimization management value of the data center is obtained based on the best table resources, poor table resources and general table resources. Specifically, the total number of poor table resources is marked as HSp, the core resource analysis value of the poor table resources is marked as STe, e = 1, 2, ..., E, E is the total number of poor table resources, the core resource analysis coefficient is set to HAw, and the formula is used Get the resource optimization management value EAtd of the data center.
6. The resource optimization management system for data centers according to claim 1, characterized in that: Based on the comparison result between the resource optimization management value and the resource optimization management threshold, determine whether to perform resource optimization management on the data center. Specifically, set the resource optimization management threshold. When the resource optimization management value of the data center is greater than or equal to the resource optimization management threshold, determine to perform resource optimization management on the data center.
7. The resource optimization management system for data centers according to claim 1, characterized in that: The resource optimization target value of the resource optimization scheme is obtained in the following way: when the resource optimization simulation model uses the resource optimization scheme to simulate resource optimization of the data center, the resource optimization management value of the data center is obtained every t time. When the resource optimization scheme completes the simulation resource optimization of the data center, the total number of resource optimization management values is marked as FCas, and all resource optimization management values are sorted in the order of acquisition time. The ratio of the previous adjacent resource optimization management value and the next adjacent resource optimization management value after sorting is calculated to obtain the resource optimization neighbor ratio value, and the resource optimization neighbor ratio threshold is set. When the resource optimization neighbor ratio value is greater than or equal to the resource optimization neighbor ratio threshold, the number of excellent resources is increased by one. When the resource optimization neighbor ratio value is less than the resource optimization neighbor ratio threshold, the number of excellent resources is increased by one, and the number of excellent resources is marked as STge, and the number of excellent resources is marked as ASkm. All resource optimization management values are summed and averaged to obtain the resource optimization management mean MPzw. The formula is used The reasonable resource optimization value YHHn of the resource optimization scheme is obtained, where ga is the resource optimization quantity coefficient, gb is the resource optimal quantity coefficient, gc is the resource excellent quantity coefficient, and gd is the optimization management mean coefficient. The reasonable resource optimization values of all historical resource optimization schemes are summed and averaged to obtain the reasonable historical optimization value GSXz. Obtain the resource optimization target value BZET.
8. The resource optimization management system for data centers according to claim 1, characterized in that: Based on the comparison result between the resource optimization target value and the resource optimization target threshold, determine whether the resource optimization plan is a resource optimization target plan, specifically: set the resource optimization target threshold, when the resource optimization target value of the resource optimization plan is greater than or equal to the resource optimization target threshold, stop generating resource optimization plans, and mark the resource optimization plan as a resource optimization target plan; when the resource optimization target value of the resource optimization plan is less than the resource optimization target threshold, mark the resource optimization plan as a resource optimization historical plan, and continue to generate the next resource optimization plan.
9. A data center-oriented resource optimization management method, applied to a data center-oriented resource optimization management system according to any one of claims 1 to 8, characterized in that: The steps include: Step 1: Regularly update the core resource analysis values of each core resource in the data center, and divide the core resources of the data center into good table resources, poor table resources, and general table resources; Step 2: Obtain a resource optimization management value of the data center according to the best meter resources, the poor meter resources, and the average meter resources, and determine whether to perform resource optimization management on the data center based on a comparison result between the resource optimization management value and the resource optimization management threshold; Step 3: After determining to perform resource optimization management on the data center, obtain the actual data set of core resource indicators and the target data set of core resource indicators of each core resource, and build a resource optimization simulation model of the data center based on the actual data set of core resource indicators and the target data set of core resource indicators of each core resource; Step 4: Control the resource optimization simulation model of the data center to perform simulation resource optimization and gradually generate multiple resource optimization solutions; Step 5: Each time a resource optimization plan is generated, the resource optimization target value of the resource optimization plan is obtained. Based on the comparison result between the resource optimization target value and the resource optimization target threshold, it is determined whether the resource optimization plan is a resource optimization target plan, and the resource optimization target plan is used to perform resource optimization management on the data center.