Data center resource management method and device

Abstract

The invention discloses a data center resource management method and device, and the method comprises the steps: carrying out the modeling of at least one data center and at least one calculation service, and obtaining a data center model of each data center and a calculation service model corresponding to each calculation service; collecting calculation service information and data center information based on the data center model and the calculation service model; determining a target optimization function of each resource optimization target based on the data center model, the calculation service model and the at least one resource optimization target; based on the calculation service information, the data center information and at least one target constraint condition, performing iterative solution on the target optimization function to obtain solved target service deployment information; and performing resource allocation management on the data center based on the target service deployment information. Through the technical scheme of the embodiment of the invention, data center resource management under the condition of considering the multi-target influence is realized, the reasonability of resource allocation is improved, and the overall performance of the data center is improved.

Description

Technical Field

[0001] This case is a divisional application of an invention patent with an application number of 202411369222.0, an application date of September 29, 2024, and an invention title of "A Data Center Resource Management Method and Device". The present invention relates to the field of computer technology, and particularly to a data center resource management method and device. Background Art

[0002] With the continuous development of the data center technology system, data center resource management has become a key technology for improving data center resource utilization, reducing operating costs, and ensuring service quality.

[0003] Currently, traditional data center resource management methods mainly focus on a single goal, such as minimizing energy consumption or maximizing resource utilization. However, traditional data center management methods often ignore the mutual influence and constraints between multiple goals, resulting in unreasonable resource allocation and affecting the overall performance of the data center. Summary of the Invention

[0004] The present invention provides a data center resource management method and device to achieve data center resource management considering the influence of multiple goals, improve the rationality of resource allocation, and improve the overall performance of the data center.

[0005] In a first aspect, an embodiment of the present invention provides a data center resource management method, including:

[0006] Model at least one data center and at least one type of computing service to obtain a data center model corresponding to each data center and a computing service model corresponding to each type of computing service;

[0007] Based on the data center model and the computing service model, collect corresponding computing service information and data center information;

[0008] Based on the data center model, the computing service model, and at least one resource optimization goal, determine a target optimization function corresponding to each resource optimization goal;

[0009] Based on the computing service information, the data center information, and at least one target constraint condition, perform iterative solution on the target optimization function to obtain the solved target service deployment information;

[0010] Based on the target service deployment information, perform resource allocation management on the data center.

[0011] In a second aspect, an embodiment of the present invention further provides a data center resource management device, including:

[0012] A model construction module for modeling at least one data center and at least one type of computing service to obtain a data center model corresponding to each data center and a computing service model corresponding to each type of computing service;

[0013] A data collection module for collecting corresponding computing service information and data center information based on the data center model and the computing service model;

[0014] An optimization function determination module for determining a target optimization function corresponding to each resource optimization target based on the data center model, the computing service model, and at least one resource optimization target;

[0015] A deployment information determination module for iteratively solving the target optimization function based on the computing service information, the data center information, and at least one target constraint condition to obtain the solved target service deployment information;

[0016] A resource allocation module for performing resource allocation management on the data center based on the target service deployment information.

[0017] The technical solution of the embodiment of the present invention, by modeling at least one data center and at least one type of computing service to obtain a data center model corresponding to each data center and a computing service model corresponding to each type of computing service, helps to understand the complexity of the data center and the computing service, and provides a basic framework for subsequent data collection and optimization solution. Based on the data center model and the computing service model, collect the corresponding computing service information and data center information, and obtain the operating status of the data center in real time, providing a basis for subsequent resource allocation. Based on the data center model, the computing service model, and at least one resource optimization target, determine the target optimization function corresponding to each resource optimization target, making the optimization process more flexible and controllable. Based on the computing service information, the data center information, and at least one target constraint condition, iteratively solve the target optimization function to obtain the solved target service deployment information, which can meet the resource management requirements in different scenarios and improve resource utilization efficiency. Based on the target service deployment information, perform resource allocation management on the data center, realize dynamic adjustment and optimal configuration of resources, and improve the operating efficiency and flexibility of the data center. Through systematic modeling, status collection, multi-objective optimization solution, and resource configuration of the data center, the efficient and dynamic management of the data center resources is realized, which helps to improve the operating efficiency of the data center, improve the rationality of resource allocation, improve the overall performance of the data center, accurately match the computing service, and thus improve the resource utilization rate and the providing efficiency and quality of the computing service.

[0018] It should be understood that the content described in this section is not intended to identify the key or important features of the embodiments of the present invention, nor is it used to limit the scope of the present invention. Other features of the present invention will become easily understood through the following description. BRIEF DESCRIPTION OF THE DRAWINGS

[0019] In order to more clearly illustrate the technical solutions in the embodiments of the present invention, the following will briefly introduce the drawings required for the description of the embodiments. Obviously, the drawings in the following description are only some embodiments of the present invention. For those of ordinary skill in the art, without creative efforts, other drawings can be obtained based on these drawings.

[0020] Figure 1 is a flowchart of a data center resource management method provided according to Embodiment 1 of the present invention;

[0021] Figure 2 is a flowchart of a data center resource management method provided according to Embodiment 2 of the present invention;

[0022] Figure 3 is a schematic structural diagram of a data center resource management device provided according to Embodiment 3 of the present invention. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0023] In order to enable those skilled in the art to better understand the solutions of the present invention, the following will clearly and completely describe the technical solutions in the embodiments of the present invention in conjunction with the drawings in the embodiments of the present invention. Obviously, the described embodiments are only a part of the embodiments of the present invention, rather than all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those of ordinary skill in the art without creative efforts shall fall within the protection scope of the present invention.

[0024] It should be noted that the terms "target", "current", etc. in the specification and claims of the present invention and the above-mentioned drawings are used to distinguish similar objects, and do not necessarily need to describe a specific order or sequence. It should be understood that such data can be interchanged under appropriate circumstances so that the embodiments of the present invention described here can be implemented in an order different from those illustrated or described here. In addition, the terms "comprising" and "having" and any variations thereof are intended to cover non-exclusive inclusion. For example, a process, method, system, product, or device that includes a series of steps or units does not necessarily have to be limited to those steps or units clearly listed, but may include other steps or units not clearly listed or inherent to these processes, methods, products, or devices.

[0025] Embodiment 1

[0026] Figure 1FIG. 0 is a flowchart of a method for managing data center resources according to Embodiment 1 of the present invention. This embodiment is applicable to the situation of managing resources in a data center. As Figure 1 shown, this method can be executed by a data center resource management device, which can be implemented in the form of hardware and / or software, and can be configured in an electronic device. As Figure 1 shown, the method specifically includes the following steps:

[0027] S110. Model at least one data center and at least one type of computing service to obtain a data center model corresponding to each data center and a computing service model corresponding to each type of computing service.

[0028] Among them, a data center can refer to a centralized computing facility (such as a computer room), which usually includes a large number of servers, storage devices, network devices, and related software and services for processing, storing, and distributing data. A computing service can refer to a service that uses the computing resources provided by a data center to perform specific tasks or provide specific functions. A data center model can be an abstract representation of the physical and logical structure of a data center. A computing service model can be an abstract representation of the function and resource requirements of a computing service.

[0029] Specifically, abstract the elements and their relationships in at least one data center to construct an abstract representation of the physical and logical structure of each data center, and obtain a data center model corresponding to each data center; model at least one type of computing service, and abstract the computing service function and resource requirements corresponding to each type of computing service to obtain a computing service model corresponding to each type of computing service. Thus, by clearly constructing the model, the complexity of the data center can be understood, which can provide a basis for subsequent data collection and optimization solution.

[0030] Exemplarily, S110 may include: modeling at least one data center to obtain a first type of multivariate mathematical expression corresponding to each data center; wherein, the dependent variable in the first type of multivariate mathematical expression is the operation performance parameter of the data center, and the independent variables are the operation resource parameters of the data center; modeling at least one type of computing service to obtain a second type of multivariate mathematical expression corresponding to each type of computing service; wherein, the dependent variable in the second type of multivariate mathematical expression is the operation performance parameter of the computing service, and the independent variables are the operation resource parameters for providing the computing service. Of course, the modeling of at least one data center and the modeling of at least one type of computing service to obtain the mathematical models corresponding to each data center and each service may also be in other forms. For example, it may only contain multiple independent variables or dependent variables, or contain both independent variables and dependent variables at the same time, but the positions of the independent variables and dependent variables are different from those in the above first type of multivariate mathematical expression and the second type of multivariate mathematical expression, or contain both independent variables and dependent variables at the same time, but the contents of the independent variables and dependent variables are different from those in the above first type of multivariate mathematical expression and the second type of multivariate mathematical expression. It can be understood that those skilled in the art can establish models from different perspectives to achieve the description of at least one data center and at least one type of computing service, and this patent does not make any limitations here.

[0031] Among them, the first type of multivariate mathematical expression may refer to a data center model for describing the relationship between the operation performance of the data center and its operation resources. The second type of multivariate mathematical expression may refer to a computing service model for describing the relationship between the operation performance of the computing service and the operation resources required therefor. The operation performance parameters may include any one or more of: service latency, service throughput, computing power matching degree, network bandwidth, computer room energy consumption, and computer room resource utilization rate. The operation resource parameters may include any one or more of: the number, model, and performance of the central processing unit and / or graphics processing unit, and may also include the number of computing service requests, etc.

[0032] Specifically, define the operating performance parameters of each data center as dependent variables (such as service throughput, computing power matching degree, network bandwidth, computer room energy consumption, and computer room resource utilization rate, etc.), and the operating resource parameters as independent variables (such as the number, model, performance, etc. of central processing units and / or graphics processing units), and construct a first type of multivariate mathematical expression corresponding to each data center based on the above independent and dependent variables to describe the complex relationship between the operating performance parameters and operating resource parameters of each data center; define the operating performance parameters of each computing service as dependent variables (such as service latency, service throughput, computing power matching degree, network bandwidth, computer room energy consumption, and computer room resource utilization rate, etc.), and the operating resource parameters required to provide the computing service as independent variables (such as the number, model, performance, etc. of central processing units and / or graphics processing units), and construct a second type of multivariate mathematical expression corresponding to each computing service based on the above independent and dependent variables to describe the complex relationship between the operating performance parameters and operating resource parameters of each computing service. Through precise model construction, the resources of the data center can be allocated and managed more reasonably, the computing services can be precisely matched, resource waste and over-allocation can be avoided, thereby improving resource utilization rate and the providing efficiency and quality of computing services.

[0033] S120. Based on the data center model and the computing service model, collect the corresponding computing service information and data center information.

[0034] Among them, the computing service information may refer to various data and metrics covering the current operating state of the computing service determined according to the computing service model. The data center information may refer to various data and metrics related to the current operating state of the overall data center and its various components determined according to the data center model. These data and metrics are used to evaluate aspects such as the performance, resource usage, health status, and resource requirements of the data center and computing services.

[0035] Specifically, based on the data center model and the computing service model, collect computing service information in real time, such as parameters of each computing service request volume, service latency, service throughput, service quality evaluation, etc., and data center information such as network bandwidth, transmission latency within and / or between computer rooms, computer room energy consumption, computing power model and quantity, computer room resource utilization rate, computing power matching degree, etc., which can ensure the accuracy and real-time nature of data collection and provide a basis for subsequent resource reallocation and optimization decision-making.

[0036] Exemplarily, S120 may include: based on the data center model and the computing service model, determine the target information corresponding to each data center and each computing service; based on the target information corresponding to each data center and each computing service, obtain the computing service information corresponding to the computing service and the current data center information corresponding to the data center.

[0037] Among them, the target information may refer to the key data and metrics that need to be collected according to the computing service model and the data center model.

[0038] Specifically, according to the first type of multivariate mathematical expressions in the data center model, determine the key performance parameters and resource usage conditions that need to be monitored and collected for each data center, that is, the target information; according to the second type of multivariate mathematical expressions corresponding to the computing service model, determine the performance metrics and resource consumption conditions that need to be tracked for each computing service, that is, the target information. Use monitoring tools and system logs to collect the computing service information corresponding to the computing service and the data center information corresponding to the data center in real time according to the target information. By monitoring and analyzing the resource usage of the data center and the computing service in real time, the demand of the computing service and the situation of resource idleness or over-allocation in the data center can be accurately reflected, so as to facilitate subsequent optimization and adjustment of resources, improve resource utilization rate, as well as the providing efficiency and quality of the computing service.

[0039] S130. Determine the target optimization function corresponding to each resource optimization goal based on the data center model, the computing service model, and at least one resource optimization goal.

[0040] Among them, the resource optimization goal may refer to the purpose that wants to be achieved by effectively managing and allocating resources in the data center and the computing service. For example, the resource optimization goal may refer to the highest resource utilization rate, the lowest power consumption, the smallest bandwidth, the lowest latency, the largest service throughput, etc.

[0041] Specifically, according to the resource optimization goal (such as minimizing energy consumption, maximizing resource utilization rate, etc.), combined with the data center model and the computing service model, construct the target optimization function corresponding to each resource optimization goal, so as to clarify the optimization direction, make the resource allocation more reasonable and efficient, and flexibly respond to different business requirements and management goals.

[0042] Exemplarily, S130 may include: obtaining at least one resource optimization goal corresponding to at least one data center and at least one computing service; taking each resource optimization goal as the dependent variable, and taking the deployment plan of at least one computing service in at least one data center as the independent variable, and constructing the target optimization function corresponding to each resource optimization goal based on the data center model and the computing service model.

[0043] Among them, the target optimization function may refer to a function used to describe and quantify how the deployment plan of the computing service in the data center affects the resource optimization goal.

[0044] Specifically, at least one resource optimization objective corresponding to at least one data center and at least one type of computing service is obtained, and the key factors affecting each resource optimization objective, that is, the deployment plan of each type of computing service in each data center, are analyzed. Each resource optimization objective is used as the dependent variable (i.e., the target value to be optimized), and the deployment plan of the computing service in the data center is used as the independent variable (i.e., the adjustable parameter). Based on the data center model and the computing service model, combined with mathematical and statistical methods, an objective optimization function that can reflect the relationship between the resource optimization objective and the deployment plan is constructed. This function is used to quantitatively describe the change of the resource optimization objective under different deployment plans. By optimizing the deployment plan of the computing service in the data center, resources can be allocated and utilized more reasonably, reducing resource idleness and waste, and improving resource utilization rate, as well as the provision efficiency and quality of the computing service.

[0045] S140. Based on the computing service information, data center information, and at least one objective constraint condition, iteratively solve the objective optimization function to obtain the solved target service deployment information.

[0046] Among them, the objective constraint condition can refer to a series of restrictive conditions that must be followed in the process of achieving the resource optimization objective. These conditions limit the solution space of the resource optimization objective and improve the feasibility and effectiveness of the resource allocation and optimization results in actual operation. The target service deployment information can refer to the specific information on how to deploy the computing service into the data center.

[0047] In one embodiment, before iteratively solving the objective optimization function, it also includes obtaining the respective parameter values and / or parameter value functions of different computing services under different deployment plans in different data centers through methods such as stress testing, simulated deployment, and actual deployment. Furthermore, the function values of the respective objective optimization functions of different computing services under different deployment plans in different data centers are determined according to the respective parameter values and / or parameter value functions.

[0048] Specifically, according to the computing service information, data center information, and objective constraint conditions, an appropriate solution and / or optimization algorithm (such as genetic algorithm, ant colony algorithm, linear programming, etc.) is used to iteratively solve the objective optimization function. During the solving process, by continuously adjusting the resource allocation plan until all constraint conditions are met and an optimal or near-optimal solution is reached, the solved target service deployment information is obtained, thereby finding the optimal or near-optimal resource allocation plan, improving resource utilization rate, reducing energy consumption and costs, and at the same time improving the provision efficiency and quality of the computing service.

[0049] S150. Based on the target service deployment information, perform resource allocation management on the data center.

[0050] Specifically, according to the solved target service deployment information, specific resource scheduling and configuration operations are implemented to adjust the resource allocation plan for the computing services provided by the data center, achieving dynamic resource allocation management for the data center, improving the overall performance and service quality of the data center. Meanwhile, the efficiency and quality of providing computing services are improved.

[0051] Exemplarily, S150 may include: based on the target service deployment information, for each computing service among at least one computing service, allocate at least one type of resource in at least one corresponding data center to meet the computing service requirements in the next time period.

[0052] Among them, the computing service requirements may refer to the specific requirements and expectations for the computing resources, storage resources, network resources, and other related resources and service capabilities corresponding to the data center.

[0053] Specifically, according to the target service deployment information, clarify the specific computing service requirements of each computing service in the next time period, and based on the computing service requirements and the resource situation of the data center, allocate the corresponding data center resources to each computing service, so as to meet the computing service requirements in the next time period. Through refined resource allocation management, it can be ensured that the computing services obtain sufficient resources when needed, guarantee the efficiency and quality of providing computing services, and at the same time avoid over-allocation and waste of resources.

[0054] It should be noted that after completing the resource allocation management for the data center, the next round of model abstraction (data center model and computing service model) can be corrected according to the management results. This correction operation can be executed regularly, for example, solved once every half hour, or can be executed in real time. Through regular or real-time model correction, it can more accurately reflect the current resource usage situation and business requirement changes of the data center. This helps to more finely adjust the resource allocation strategy, reduce resource waste, and improve the overall resource utilization efficiency as well as the efficiency and quality of providing computing services.

[0055] The technical solution of the embodiment of the present invention models at least one data center and at least one type of computing service to obtain a data center model corresponding to each data center and a computing service model corresponding to each type of computing service, thereby helping to understand the complexity of the data center and the computing service, and providing a basic framework for subsequent data collection and optimization solving. Based on the data center model and the computing service model, the corresponding computing service information and data center information are collected to obtain the operating status of the data center in real time, providing a basis for subsequent resource allocation. Based on the data center model, the computing service model, and at least one resource optimization goal, the target optimization function corresponding to each resource optimization goal is determined, making the optimization process more flexible and controllable. Based on the computing service information, the data center information, and at least one target constraint condition, the target optimization function is iteratively solved to obtain the solved target service deployment information, which can meet the resource management requirements in different scenarios, improve the resource utilization efficiency, as well as the providing efficiency and quality of the computing service. Based on the target service deployment information, resource allocation management is performed on the data center to achieve dynamic adjustment and optimal configuration of resources, improving the operating efficiency and flexibility of the data center. Through systematic modeling, status collection, multi-objective optimization solving, and resource configuration of the data center, efficient and dynamic management of the data center resources is achieved, which helps to improve the operating efficiency of the data center, improve the rationality of resource allocation, improve the overall performance of the data center, and at the same time improve the providing efficiency and quality of the computing service.

[0056] It should be noted that a data center often includes many computer rooms (data centers), and the computer rooms may be spread across different regions to meet the access needs of users in various places. These computer rooms deployed in different places are for improving the access experience of local users, for business backup of the computer rooms, or for the need of cluster business expansion. No matter what services are deployed, when doing cross-computer room operations, the possible problems that may be faced should be fully considered to avoid making mistakes during cross-computer room deployment.

[0057] 1. Taking the bandwidth issue as an example, to connect two or more computer rooms in different regions, it is often necessary to go through the wide area network. It is necessary to rent bandwidth from the operator and establish a VPN tunnel between the two computer rooms to achieve the connection. The size of this bandwidth should be based on the actual needs of the computer rooms and also consider the bandwidth that the operator can provide. In principle, if local computer room forwarding can be performed, try not to go through cross-computer room to reduce the traffic burden of cross-computer room, which can reduce the cost of renting the operator's network bandwidth. When doing cross-computer room service deployment, first consider whether the bandwidth at the computer room entrance and exit can meet the requirements. If it cannot meet the requirements, it is a veto, and such cross-computer room services cannot be built and deployed.

[0058] 2. Taking security issues as an example, within a data center, various software and hardware security protection measures can be deployed to ensure data security. However, for data interaction between computer rooms, this data is no longer under the control of the computer rooms. When transmitted over a wide area network, the security of the data center computer room is difficult to play a role. At this time, some of the data may have an outer IP header added and be encapsulated, but the inner message may still be intercepted and restored by someone. Some may be encrypted on the security devices of the sending and receiving computer rooms, but these encryption algorithms are not indecipherable. As long as someone masters the deciphering method, the data may still be stolen during the transmission process. There are security vulnerabilities in the various wide area network protocols themselves. In case of an attack and being exploited, the data will be leaked, and the resulting losses may be fatal. Therefore, try to transfer computational data or intermediate process data between computer rooms. For some data involving personal information or business secrets, try not to transfer them repeatedly in cross-computer room services.

[0059] 3. Taking latency issues as an example, the latency of data transmission is also an important indicator. The latency within a data center is usually very low, at the microsecond level. When data is transmitted between data centers, affected by the latency of the cross-data center network, the latency will be relatively high, usually at the millisecond level. The data transmission speed and latency of an Internet data center depend on multiple factors, including network bandwidth, network topology, and the device performance within the data center, etc. Generally speaking, the data transmission speed within a data center can be very fast, usually transmitting data at a speed of gigabit or even ten gigabit. The internal network communication latency is generally about 300 us (0.3 ms). When data is transmitted between data centers, it will be restricted by the cross-data center network bandwidth and is generally slower than the transmission speed within the data center. The cross-computer room communication latency may be as high as 50000 us (50 ms). When conducting business deployment, the cross-computer room latency issue should be fully considered.

[0060] Embodiment 2

[0061] Figure 2 The flowchart of a data center resource management method provided by Embodiment 2 of the present invention. Based on the above embodiments, this embodiment optimizes the step of "iteratively solving the target optimization function based on the computing service information, data center information, and at least one target constraint condition to obtain the solved target resource deployment information". The explanations of the same or corresponding terms as those in the above embodiments will not be repeated here.

[0062] See Figure 2 , another data center resource management method provided by this embodiment specifically includes the following steps:

[0063] S210. Model at least one data center and at least one type of computing service to obtain a data center model corresponding to each data center and a computing service model corresponding to each type of computing service.

[0064] S220. Based on the data center model and the computing service model, collect the corresponding computing service information and data center information.

[0065] S230. Based on the data center model, the computing service model, and at least one resource optimization goal, determine a target optimization function corresponding to each resource optimization goal.

[0066] S240. Obtain at least one target constraint condition.

[0067] Specifically, the target constraint condition can refer to a series of restrictive conditions that must be followed during the process of achieving the resource optimization goal. These conditions limit the solution space of the resource optimization goal and improve the feasibility and effectiveness of the optimization result in actual operation.

[0068] It should be noted that the target constraint condition can be in the same dimension as the target optimization function. For example, the power consumption of each data center does not exceed a preset value, and at the same time, a sub-constraint goal is to minimize the power consumption of the data center (the power consumption is different when the computer room load is light and heavy); the cross-computer room network bandwidth is limited to below 100 Gb / s, and at the same time, a sub-constraint goal is to minimize the cross-computer room network bandwidth. And so on. Among them, the power consumption not exceeding the constraint value and the bandwidth not exceeding 100 G are strong constraints, and the sub-constraint goal can be a weak constraint.

[0069] Exemplarily, S240 may further include: obtaining a constraint level corresponding to at least one target constraint condition.

[0070] Specifically, each target constraint condition has a corresponding constraint level, which is used to indicate the importance and priority of the condition in the process of solution and optimization.

[0071] S250. Based on the computing service information, the data center information, and at least one target constraint condition, perform a preliminary solution to the target optimization function to obtain at least one original resource deployment information after the preliminary solution.

[0072] Among them, the original resource deployment information can refer to one or more groups of resource allocation schemes obtained after the preliminary solution of the target optimization function, and these schemes are used to illustrate how to allocate computing resources and other related resources to different computing services.

[0073] Specifically, according to the computing service information, the data center information, and the target constraint condition, perform a preliminary solution to the target optimization function through methods such as linear programming to obtain at least one original resource deployment information, providing a basis for subsequent optimization iterations.

[0074] Exemplarily, according to the current various types of computing service requirements and the data center information corresponding to each data center, when determining which data center each service is deployed in, and even being able to be precise to certain computing power resources in a certain data center, the value of each independent variable and the value of each objective optimization function are obtained to acquire at least one original resource deployment information. For example, at a certain moment, there are four computing requirements, namely 1, 2, 3, and 4, and there are four data centers, A, B, C, and D. If the original resource deployment information is deployed according to the 1A, 2B, 3C, 4D plan, then the values of variables such as power consumption, latency, and bandwidth requirements can be determined, and the values of each objective function at this time can also be determined; if the original resource deployment information is deployed according to the 1B, 2A, 3D, 4C plan, then the values of variables such as power consumption, latency, and bandwidth requirements can be determined, and the values of each objective function at this time can also be determined. The solution of each objective function is a potential original solution of the multi-data center service deployment plan based on multi-objective constrained programming, that is, the potential original resource deployment information. Then, by determining whether it meets at least one objective constraint condition, the original solution, that is, the original resource deployment information, is determined.

[0075] Exemplarily, S250 may include: based on the computing service information, data center information, at least one objective constraint condition, and the constraint level corresponding to the objective constraint condition, performing a preliminary solution to the objective optimization function to obtain at least one original resource deployment information after the preliminary solution.

[0076] Specifically, based on the computing service information, data center information, and at least one objective constraint condition, the objective optimization function is preliminarily solved by means such as greedy, linear programming, enumeration, and exhaustive search to obtain the original resource deployment information corresponding to each objective optimization function after the preliminary solution. By considering the objective constraint condition and its constraint level, it can be ensured that the resource deployment plan is more in line with the actual requirements and priorities, and the rationality of resource deployment is improved.

[0077] Exemplarily, based on the computing service information, data center information, at least one objective constraint condition, and the constraint level corresponding to the objective constraint condition, performing a preliminary solution to the objective optimization function to obtain at least one original resource deployment information after the preliminary solution further includes: if there is no original resource deployment information after the preliminary solution to the objective optimization function, or the number of original resource deployment information is lower than the preset value, then a downgrade adjustment is performed on at least one constraint level corresponding to the objective constraint condition; based on the computing service information, data center information, at least one objective constraint condition, and the adjusted objective constraint condition corresponding to the objective constraint condition, the objective optimization function is re-preliminarily solved to obtain at least one original resource deployment information after the preliminary solution.

[0078] Specifically, if there is no original resource deployment information, or the number of original resource deployment information is lower than a preset value (such as not meeting the minimum solution set size requirement), it is determined that the current constraint condition is too strict. Select the target constraint condition corresponding to the target optimization function without original resource deployment information, and downgrade the corresponding constraint level. This usually means relaxing the restriction scope of these constraint conditions. Based on the computing service information, data center information, at least one target constraint condition, and the adjusted target constraint condition corresponding to the target constraint condition, re-initialize the solution of the target optimization function to obtain at least one original resource deployment information after the initial solution. Check the new solution result to confirm whether the required original resource deployment information is obtained. If the requirements are still not met after re-solving, the constraint level can be further adjusted, and the above process can be repeated. Through iterative adjustment and solution, gradually approach the optimal solution set that meets all conditions. By downgrading the constraint level, the difficulty of solving can be reduced, and the success rate of finding a feasible solution can be increased.

[0079] Exemplarily, if there is no original resource deployment information, adjust and downgrade the target constraint condition and try to re-solve. At this time, the target constraint condition can be constrained and downgraded through a preset adjustment rule, which usually occurs when the target optimization function cannot obtain any solution. For example, when the first solution is attempted, a strong constraint may be that all of computing service requirement 1 is deployed in data center A. At this time, according to the collected data, requirement 1 is greater than the service capacity of computer room A. At this time, any deployment plan under this strong constraint condition cannot meet the target constraint condition, that is, the target optimization function has no solution. At this time, the strong constraint must be downgraded to a weak constraint through constraint adjustment, that is, requirement 1 is preferentially deployed in data center A, and the demand exceeding the service capacity can provide services across data centers. At this time, there may be an optimal solution.

[0080] S260. Based on the multi-objective optimization algorithm, optimize and iterate at least one original resource deployment information corresponding to the target optimization function to obtain the target resource deployment information after optimization and iteration.

[0081] Among them, the multi-objective optimization algorithm can refer to an algorithm used to simultaneously process multiple optimization objectives. The goal of this algorithm is to find the best trade-off solution among multiple conflicting or related objectives, so that all objectives can be satisfied to a certain extent, rather than just optimizing one of the objectives.

[0082] Specifically, since resource deployment issues often involve multiple optimization objectives (such as cost minimization, performance maximization, maximum resource utilization, etc.), a multi-objective optimization algorithm needs to be used to find the optimal solution. The multi-objective optimization algorithms that can be adopted can include any one or more heuristic algorithms such as genetic algorithms, particle swarm optimization algorithms, ant colony algorithms, and greedy algorithms. In one embodiment, when multiple multi-objective optimization algorithms are adopted, weights can be set for the optimal solutions obtained by optimizing each multi-objective optimization algorithm respectively, and then the calculation service deployment solutions of multiple optimal solutions are multiplied by the corresponding weights to obtain the final solution. Through multiple iterative optimizations of the initially obtained original resource deployment information by the multi-objective optimization algorithm, the resource allocation scheme is continuously adjusted to approach the optimal solution. During the iteration process, the algorithm will gradually improve the effect of resource deployment according to the feedback of the objective optimization function and the limitations of the objective constraint conditions, and obtain the target resource deployment information after optimization iteration. Through optimization iteration, it can be ensured that resources are more reasonably allocated and utilized, resource waste is reduced, and the overall resource utilization efficiency as well as the provision efficiency and quality of computing services are improved.

[0083] S270. Based on the target service deployment information, perform resource allocation management for the data center.

[0084] The technical solution of the embodiment of the present invention can ensure that the optimization process always proceeds in the direction of meeting actual requirements by obtaining at least one target constraint condition. Based on the computing service information, data center information, and at least one target constraint condition, perform preliminary solution for the target optimization function to obtain at least one original resource deployment information after preliminary solution, which can verify the rationality of the optimization function and the effectiveness of the solution algorithm, reduce the risks in the subsequent optimization iteration process, and provide a basis for the subsequent multi-objective optimization iteration. Based on the multi-objective optimization algorithm, perform optimization iteration on at least one original resource deployment information corresponding to the target optimization function to obtain the target resource deployment information after optimization iteration, which can find a balance among multiple optimization objectives, thereby more reasonably allocate resources and improve resource utilization. By clarifying multiple target constraint conditions, preliminary solution, and iteration of the multi-objective optimization algorithm, the data center resource deployment problem can be effectively solved, the efficient and reasonable configuration of data center resources is realized, the overall performance and resource utilization of the data center are improved, and at the same time, the provision efficiency and quality of computing services are improved.

[0085] Exemplarily, the above-mentioned greedy algorithm (also known as the greedy algorithm) means that when solving a problem, it always makes the best choice at present. That is to say, it does not consider the overall optimum, and the solution obtained by the algorithm is a local optimum in a certain sense. The greedy algorithm does not obtain the overall optimum for all problems. The key is the selection of the greedy strategy.

[0086] The basic idea of the greedy algorithm is to proceed step by step from a certain initial solution of the problem. According to a certain optimization measure, a local optimal solution should be obtained at each step. Only one data is considered at each step, and its selection should meet the conditions of local optimization. If the next data and the partial optimal solution together are no longer a feasible solution, the data is not added to the partial solution until all data are enumerated or no more data can be added and the algorithm stops.

[0087] The greedy algorithm generally proceeds as follows:

[0088] ① Establish a mathematical model to describe the problem.

[0089] ② Divide the problem to be solved into several sub-problems.

[0090] ③ Solve each sub-problem to obtain the local optimal solution of the sub-problem.

[0091] ④ Combine the local optimal solutions of the sub-problems into a solution to the original problem.

[0092] The greedy algorithm is a simpler and faster design technique for some problems of finding the optimal solution. The characteristic of the greedy algorithm is to proceed step by step, often making the optimal choice based on a certain optimization measure according to the current situation, without considering various possible overall situations, saving the large amount of time that must be consumed to exhaust all possibilities to find the optimal solution. The greedy algorithm adopts a top-down approach and makes successive greedy choices iteratively. Each time a greedy choice is made, the problem to be solved is reduced to a sub-problem with a smaller scale. Through each step of greedy choice, an optimal solution to the problem can be obtained. Although a local optimal solution must be obtained at each step, the resulting global solution is sometimes not necessarily optimal, so the greedy algorithm does not backtrack.

[0093] Linear programming (LP) is an important branch in operations research that has been studied earlier, developed faster, widely applied, and has relatively mature methods. It is a mathematical method to assist people in scientific management and is a mathematical theory and method for studying the extreme value problems of linear objective functions under linear constraint conditions.

[0094] Linear programming is an important branch of operations research and is widely applied in military operations, economic analysis, business management, engineering technology, etc. It provides a scientific basis for making optimal decisions to rationally utilize limited resources such as manpower, material resources, and financial resources.

[0095] Introduction to Linear Programming

[0096] (1) List the constraint conditions and the objective function

[0097] (2) Draw the feasible region represented by the constraint conditions

[0098] (3) Find the optimal solution and optimal value of the objective function within the feasible region.

[0099] Embodiment III

[0100] Figure 3 The following is a schematic structural diagram of a data center resource management device provided in Embodiment III of the present invention. As Figure 3 shown, the device includes: a model construction module 310, a data collection module 320, an optimization function determination module 330, a deployment information determination module 340, and a resource allocation module 350.

[0101] Among them, the model construction module 310 is used to model at least one data center and at least one type of computing service to obtain a data center model corresponding to each data center and a computing service model corresponding to each type of computing service;

[0102] The data collection module 320 is used to collect corresponding computing service information and data center information based on the data center model and the computing service model;

[0103] The optimization function determination module 330 is used to determine a target optimization function corresponding to each resource optimization target based on the data center model, the computing service model, and at least one resource optimization target;

[0104] The deployment information determination module 340 is used to iteratively solve the target optimization function based on the computing service information, the data center information, and at least one target constraint condition to obtain the solved target service deployment information;

[0105] The resource allocation module 350 is used to perform resource allocation management on the data center based on the target service deployment information.

[0106] The technical solution of this embodiment models at least one data center and at least one type of computing service to obtain a data center model corresponding to each data center and a computing service model corresponding to each type of computing service, thereby helping to understand the complexity of data centers and computing services and providing a basic framework for subsequent data collection and optimization solution. Based on the data center model and the computing service model, the corresponding computing service information and data center information are collected to obtain the operating status of the data center in real time, providing a basis for subsequent resource allocation. Based on the data center model, the computing service model, and at least one resource optimization goal, the target optimization function corresponding to each resource optimization goal is determined, making the optimization process more flexible and controllable. Based on the computing service information, the data center information, and at least one target constraint condition, the target optimization function is iteratively solved to obtain the solved target service deployment information, which can meet the resource management requirements in different scenarios and improve resource utilization efficiency. Based on the target service deployment information, resource allocation management is performed on the data center to achieve dynamic adjustment and optimal configuration of resources, improving the operating efficiency and flexibility of the data center. Through systematic modeling, status collection, multi-objective optimization solution, and resource configuration of the data center, efficient and dynamic management of data center resources is achieved, which helps to improve the operating efficiency of the data center, improve the rationality of resource allocation, and improve the overall performance of the data center.

[0107] Optionally, the model construction module 310 is specifically configured to: model at least one data center to obtain a first type of multivariate mathematical expression corresponding to each data center; wherein, the dependent variable in the first type of multivariate mathematical expression is the operating performance parameter of the data center, and the independent variable is the operating resource parameter of the data center; model at least one type of computing service to obtain a second type of multivariate mathematical expression corresponding to each type of computing service; wherein, the dependent variable in the second type of multivariate mathematical expression is the operating performance parameter of the computing service, and the independent variable is the operating resource parameter for providing the computing service.

[0108] Optionally, the operating performance parameters include any one or more of: service latency, service throughput, computing power matching degree, network bandwidth, computer room energy consumption, and computer room resource utilization rate; the operating resource parameters include any one or more of: the number, model, and performance of central processing units and / or graphics processing units.

[0109] Optionally, the data collection module 320 is specifically configured to: based on the data center model and the computing service model, determine the target information corresponding to each data center and each computing service; based on the target information corresponding to each data center and each computing service, obtain the computing service information corresponding to the computing service and the current data center information corresponding to the data center.

[0110] Optionally, the optimization function determination module 330 is specifically configured to: obtain at least one resource optimization objective corresponding to the at least one data center and at least one computing service; use each resource optimization objective as a dependent variable, and use the deployment scheme of the at least one computing service in the at least one data center as an independent variable, and construct a target optimization function corresponding to each resource optimization objective based on the data center model and the computing service model.

[0111] Optionally, the deployment information determination module 340 includes:

[0112] A constraint condition acquisition unit, configured to acquire at least one target constraint condition;

[0113] An original information acquisition unit, configured to perform a preliminary solution on the target optimization function based on the computing service information, the data center information, and at least one target constraint condition, and obtain at least one original resource deployment information after the preliminary solution;

[0114] A target information acquisition unit, configured to perform an optimization iteration on at least one original resource deployment information corresponding to the target optimization function based on a multi-objective optimization algorithm, and obtain target resource deployment information after the optimization iteration.

[0115] Optionally, the constraint condition acquisition unit is specifically configured to: acquire a constraint level corresponding to at least one of the target constraint conditions.

[0116] Optionally, the original information acquisition unit is specifically configured to: perform a preliminary solution on the target optimization function based on the computing service information, the data center information, at least one target constraint condition, and the constraint level corresponding to the target constraint condition, and obtain at least one original resource deployment information after the preliminary solution.

[0117] Optionally, the original information acquisition unit is specifically configured to: if there is no original resource deployment information after a preliminary solution of the target optimization function, or the number of the original resource deployment information is lower than a preset value, perform a downgrade adjustment on at least one constraint level corresponding to the target constraint condition; perform a re-preliminary solution on the target optimization function based on the computing service information, the data center information, at least one target constraint condition, and the adjusted target constraint condition corresponding to the target constraint condition, so as to obtain at least one original resource deployment information after the preliminary solution.

[0118] Optionally, the resource allocation module 350 is specifically configured to: based on the target service deployment information, allocate at least one resource in at least one of the at least one data centers to each computing service in the at least one computing service, so as to meet the computing service requirements in the next time period.

[0119] The data center resource management device provided by the embodiments of the present invention can execute the data center resource management method provided by any embodiment of the present invention, and has the corresponding functional modules and beneficial effects for executing the method.

[0120] Those of ordinary skill in the art should understand that the above-mentioned modules or steps of the present invention can be implemented by a general-purpose computing device. They can be concentrated on a single computing device or distributed on a network composed of multiple computing devices. Optionally, they can be implemented by program codes executable by a computer device, so that they can be stored in a storage device and executed by the computing device, or they can be separately fabricated into individual integrated circuit modules, or multiple modules or steps among them can be fabricated into a single integrated circuit module for implementation. Thus, the present invention is not limited to any specific combination of hardware and software.

[0121] Note that the above is only the preferred embodiment of the present invention and the applied technical principle. Those skilled in the art will understand that the present invention is not limited to the specific embodiments described herein. Various obvious changes, re-adjustments, and substitutions can be made by those skilled in the art without departing from the protection scope of the present invention. Therefore, although the present invention has been described in detail through the above embodiments, the present invention is not limited to the above embodiments. Without departing from the concept of the present invention, more other equivalent embodiments can be included, and the scope of the present invention is determined by the scope of the appended claims.

Claims

1. A data center resource management method, characterized in that: include: Modeling at least one data center and at least one computing service to obtain a data center model corresponding to each data center and a computing service model corresponding to each computing service; Based on the data center model and the computing service model, collecting corresponding computing service information and data center information; Determine a target optimization function corresponding to each resource optimization target based on the data center model, the computing service model and at least one resource optimization target; Iteratively solving the target optimization function based on the computing service information, the data center information, and at least one target constraint condition to obtain solved target service deployment information, including: Obtain at least one target constraint; Based on the computing service information, the data center information and at least one target constraint condition, a target optimization function is preliminarily solved to obtain at least one original resource deployment information after the preliminarily solution; Based on a multi-objective optimization algorithm, at least one original resource deployment information corresponding to the target optimization function is optimized and iterated to obtain target resource deployment information after optimization and iteration; Based on the target service deployment information, resource allocation management is performed on the data center.

2. The method according to claim 1, characterized in that The obtaining of at least one target constraint condition comprises: Obtaining a constraint level corresponding to at least one of the target constraint conditions; The preliminarily solving the target optimization function based on the computing service information, the data center information and the at least one target constraint condition to obtain at least one original resource deployment information after the preliminarily solution includes: Based on the computing service information, the data center information, at least one target constraint condition and the constraint level corresponding to the target constraint condition, the target optimization function is preliminarily solved to obtain at least one original resource deployment information after the preliminary solution.

3. The method according to claim 2, characterized in that The preliminarily solving the target optimization function based on the computing service information, the data center information, the at least one target constraint condition and the constraint level corresponding to the target constraint condition to obtain at least one original resource deployment information after the preliminarily solution includes: If there is no original resource deployment information after the target optimization function is initially solved, or the amount of the original resource deployment information is lower than a preset value, downgrading at least one constraint level corresponding to the target constraint condition; Based on the computing service information, the data center information, at least one target constraint and the adjusted target constraint corresponding to the target constraint, the target optimization function is re-preliminarily solved to obtain at least one original resource deployment information after preliminary solution.

4. The method according to claim 1, characterized in that: The optimizing and iterating at least one original resource deployment information corresponding to the target optimization function based on the multi-objective optimization algorithm to obtain the target resource deployment information after the optimization and iteration includes: Determine the optimal solution and weight obtained by iteratively optimizing at least one original resource deployment information corresponding to the objective optimization function using multiple multi-objective optimization algorithms; The sum of the products of the optimal solutions corresponding to multiple multi-objective optimization algorithms and the weights is taken as the target resource deployment information.

5. The method according to claim 1, characterized in that The step of modeling at least one data center and at least one computing service to obtain a data center model corresponding to each data center and a computing service model corresponding to each computing service includes: Modeling at least one data center to obtain a first-type multivariate mathematical expression corresponding to each data center; wherein the dependent variable in the first-type multivariate mathematical expression is an operating performance parameter of the data center, and the independent variable is an operating resource parameter of the data center; Model at least one computing service to obtain a second type of multivariate mathematical expression corresponding to each computing service; wherein the dependent variable in the second type of multivariate mathematical expression is an operating performance parameter of the computing service, and the independent variable is an operating resource parameter providing the computing service.

6. The method according to claim 5, characterized in that The operating performance parameters include: any one or more of service delay, service throughput, computing power matching, network bandwidth, computer room energy consumption and computer room resource utilization; the operating resource parameters include: any one or more of the number, model and performance of the central processing unit and / or graphics processing unit.

7. The method according to claim 1, characterized in that The collecting corresponding computing service information and data center information based on the data center model and the computing service model includes: Based on the data center model and the computing service model, determining target information corresponding to each data center and each computing service; Based on the target information corresponding to each data center and each computing service, computing service information corresponding to the computing service and current data center information corresponding to the data center are acquired.

8. The method according to claim 1, characterized in that: The determining, based on the data center model, the computing service model and at least one resource optimization goal, a target optimization function corresponding to each resource optimization goal includes: Obtaining at least one resource optimization target corresponding to the at least one data center and the at least one computing service; Taking each resource optimization target as a dependent variable and the deployment scheme of the at least one computing service in the at least one data center as an independent variable, a target optimization function corresponding to each resource optimization target is constructed based on the data center model and the computing service model.

9. The method according to claim 1, characterized in that: Before iteratively solving the target optimization function, the method further includes: Obtain various parameter values ​​and / or parameter value functions of different computing services under deployment schemes in different data centers; According to each parameter value and / or parameter value function, the function value of each objective optimization function under the deployment scheme of different computing services in different data centers is determined.

10. A data center resource management device, characterized in that: include: A model building module, used to model at least one data center and at least one computing service, and obtain a data center model corresponding to each data center and a computing service model corresponding to each computing service; A data collection module, used to collect corresponding computing service information and data center information based on the data center model and the computing service model; An optimization function determination module, configured to determine a target optimization function corresponding to each resource optimization target based on the data center model, the computing service model and at least one resource optimization target; A deployment information determination module, configured to iteratively solve the target optimization function based on the computing service information, the data center information and at least one target constraint condition to obtain the solved target service deployment information; The deployment information determination module includes: A constraint condition acquisition unit, used to acquire at least one target constraint condition; An original information acquisition unit, configured to preliminarily solve the target optimization function based on the computing service information, the data center information and at least one target constraint condition, and obtain at least one original resource deployment information after the preliminarily solution; The target information acquisition unit is used to optimize and iterate at least one original resource deployment information corresponding to the target optimization function based on a multi-objective optimization algorithm to obtain the target resource deployment information after optimization and iteration. A resource allocation module is used to manage resource allocation for the data center based on the target service deployment information.

Images