A data center management method, system and storage medium

Abstract

The application discloses a data center management method and system and a storage medium, and relates to the technical field of data processing. The method comprises the following steps: acquiring power data of a data center server, and calculating server energy consumption; acquiring cooling data of the data center server; establishing a target function according to the cooling data and the server energy consumption; acquiring the utilization rate and the task amount of the data center server in real time; solving the target function by using a particle swarm algorithm; and performing task allocation on the data center according to the solution of the target function. The power data comprises processor power data, memory power data, disk power data and interface power data. The application models the server energy consumption according to the power data of the data center server, considers the task amount and the cooling capacity of the server, models and solves the total energy consumption of the server, obtains a task allocation scheme in which the total energy consumption is the lowest and the cooling of each server is sufficient, and reduces the energy consumption of the data center.

Description

Technical Field

[0001] This invention relates to the field of data processing technology, and more specifically to a data center management method, system, and storage medium. Background Technology

[0002] With the rapid development of cloud computing, data centers have become an important part of the information society. However, data centers consume a lot of energy during operation, resulting in high operating costs and low system reliability. Therefore, it is necessary to manage data centers and reduce their energy consumption.

[0003] Existing data center energy reduction solutions optimize data center management at a single level, such as reducing the number of servers deployed and handling tasks with a smaller number of servers. However, with a large number of tasks piling up on a small number of servers, the servers are heavily loaded, and prolonged operation degrades the system's computing performance, leading to slow task processing and still resulting in high energy consumption in the data center. Summary of the Invention

[0004] The purpose of this invention is to provide a data center management method, system, and storage medium. This invention models the energy consumption of servers based on the power data of data center servers, considers the workload and cooling capacity of servers, and models and solves the total energy consumption of servers to obtain a task allocation scheme with the lowest total energy consumption while ensuring sufficient cooling for each server, thereby reducing the energy consumption of the data center.

[0005] The objective of this invention is achieved through the following technical means:

[0006] In a first aspect, the present invention provides a data center management method, comprising the following steps:

[0007] Obtain power data from data center servers and calculate server energy consumption;

[0008] Obtain the cooling data of the data center server;

[0009] Based on the cooling data and the server energy consumption, establish an objective function;

[0010] Real-time acquisition of the utilization rate and workload of the data center servers;

[0011] The objective function is solved using the particle swarm optimization algorithm.

[0012] Based on the solution of the objective function, tasks are allocated to the data center;

[0013] The power data includes: processor power data, memory power data, disk power data, and interface power data.

[0014] Preferably, the step of obtaining power data of the data center server and calculating server energy consumption includes the following steps:

[0015] Obtain the processor's full-load power and idle power;

[0016] Calculate the processor's energy consumption based on the processor's full-load power and idle power.

[0017] Obtain the memory read power, write power, and refresh power;

[0018] Calculate the memory power consumption based on the memory read power, the memory write power, and the memory refresh power;

[0019] Obtain the disk's read power, write power, and idle power;

[0020] Calculate disk energy consumption based on the disk's read power, write power, and idle power.

[0021] Get the task size and bandwidth of the interface;

[0022] Calculate the interface power consumption based on the task size and bandwidth of the interface;

[0023] The server energy consumption is calculated based on the processor energy consumption, the memory energy consumption, the disk energy consumption, and the interface energy consumption.

[0024] Preferably, the formula for calculating the server's energy consumption is as follows:

[0025] ,

[0026] ,

[0027] ,

[0028] ,

[0029] ,

[0030] in, For server power consumption, As the first weight, For processor power consumption, As the second weight, For memory power consumption, As the third weight, For disk power consumption, As the fourth weight, For interface power consumption, This represents the processor's full load power. The processor's idle power, For processor utilization, For memory read power, The power consumption for memory writes. For memory refresh power, This refers to the disk's read power. This refers to the disk's write power. This refers to the idle power of the disk. The first performance factor for the interface. The task size of the interface. The bandwidth of the interface. This is the second performance factor for the interface.

[0031] Preferably, establishing the objective function based on the cooling data and the server energy consumption includes the following steps:

[0032] Calculate the cooling efficiency of the data center server based on the cooling data.

[0033] Calculate the total energy consumption of the data center server based on the cooling efficiency and the server energy consumption;

[0034] The objective function is established with the goal of minimizing the total energy consumption of the data center servers.

[0035] Preferably, the total energy consumption of the data center server is calculated using the following formula:

[0036] ,

[0037] ,

[0038] in, This represents the total energy consumption of the data center servers. For server power consumption, For cooling energy consumption, For runtime, For cooling efficiency;

[0039] The objective function is expressed as follows:

[0040] ,

[0041] ,

[0042] ,

[0043] ,

[0044] in, Let be the objective function. For the first Total energy consumption of each server For the number of servers, For the first The utilization rate of each server For the first Cooling energy consumption of a single server For cooling threshold, For the first Task size per server This refers to the workload.

[0045] Preferably, the objective function is solved using a particle swarm optimization algorithm, including the following steps:

[0046] Randomly initialize the particles to obtain the initial population;

[0047] Calculate the fitness value of the particles;

[0048] Update the optimal fitness value of the individual and the optimal fitness value of the population based on the fitness value of the particles.

[0049] The velocity and position of the particle are updated based on the optimal fitness value of the individual and the optimal fitness value of the population.

[0050] The velocity and position of the particle are repeatedly updated until the maximum number of iterations is reached.

[0051] Output the solution to the objective function to obtain the task allocation amount for the data center server;

[0052] The formula for calculating the fitness value is as follows:

[0053] ,

[0054] in, For fitness value, For the first Total energy consumption of each server This refers to the number of servers.

[0055] Preferably, updating the velocity and position of the particle based on the optimal fitness value of the individual and the optimal fitness value of the population includes the following steps:

[0056] The velocity of the particle is updated based on the optimal fitness value of the individual and the optimal fitness value of the population to obtain the first update velocity;

[0057] The position of the particle is updated according to the first update rate to obtain the first updated position;

[0058] Calculate the update fitness value of the particle based on the first update position;

[0059] The fitness difference is obtained by subtracting the updated fitness value from the individual's optimal fitness value.

[0060] When the fitness difference is greater than or equal to a preset difference, the first update speed is taken as the speed of the particle, and the first update position is taken as the position of the particle.

[0061] When the fitness difference is less than the preset difference, the first update speed and the first update position are updated using a mutation algorithm to obtain a second update speed and a second update position. The second update speed is used as the speed of the particle, and the second update position is used as the position of the particle.

[0062] Secondly, the present invention provides a data center management system that applies the above-mentioned data center management method, including: a server energy consumption calculation module, a data center cooling module, an objective function establishment module, a real-time data monitoring module, an objective function solving module, and a task allocation module;

[0063] The server energy consumption calculation module is used to obtain power data of the data center server and calculate the server energy consumption.

[0064] The data center cooling module is used to acquire cooling data from the data center servers;

[0065] The objective function establishment module is used to establish an objective function based on the cooling data and the server energy consumption;

[0066] The real-time data monitoring module is used to obtain the utilization rate and workload of the data center server in real time.

[0067] The objective function solving module is used to solve the objective function using a particle swarm optimization algorithm;

[0068] The task allocation module is used to allocate tasks to the data center based on the solution of the objective function.

[0069] Thirdly, the present invention provides an electronic device including a processor and a memory, the memory being used to store computer program code, the computer program code including computer instructions, and when the processor executes the computer instructions, the electronic device executes the aforementioned data center management method.

[0070] Fourthly, the present invention provides a computer-readable storage medium storing a computer program, the computer program including program instructions, which, when executed by a processor of an electronic device, cause the processor to perform the aforementioned data center management method.

[0071] Compared with the prior art, the beneficial effects of the present invention are as follows:

[0072] This invention models server energy consumption based on power data of data center servers, takes into account server workload and cooling capacity, models and solves the total energy consumption of servers, and obtains a task allocation scheme with the lowest total energy consumption while ensuring sufficient cooling for each server, thereby reducing the energy consumption of data centers.

[0073] This invention uses the particle swarm optimization algorithm to solve the objective function. When it gets stuck in a local optimum, it uses a mutation algorithm to update the position and velocity of the particles, which improves the solution speed while ensuring the quality of the solution, obtains the optimal task allocation scheme, and reduces the energy consumption of the data center. Attached Figure Description

[0074] The accompanying drawings, which are incorporated in and form part of this specification, illustrate embodiments consistent with the invention and, together with the description, serve to explain the principles of the invention.

[0075] To more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, for those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0076] Figure 1 This is a flowchart illustrating a data center management method provided in this embodiment;

[0077] Figure 2 This is a schematic diagram of step S1 in this embodiment, which involves obtaining power data of a data center server and calculating server energy consumption.

[0078] Figure 3 The flowchart for step S3 in this embodiment, which establishes the objective function based on cooling data and server energy consumption, is shown below.

[0079] Figure 4 This is a flowchart illustrating step S5 of the embodiment, which uses the particle swarm optimization algorithm to solve the objective function.

[0080] Figure 5The flowchart of step S54 provided in this embodiment, which updates the velocity and position of particles based on the optimal fitness value of an individual and the optimal fitness value of the group, is shown below.

[0081] Figure 6 This is a schematic diagram of the structure of a data center management system provided in this embodiment;

[0082] Figure 7 This is a schematic diagram of the structure of an electronic device provided in this embodiment. Detailed Implementation

[0083] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only a part of the embodiments of the present invention, and not all of them. Based on the embodiments of the present invention, all other embodiments obtained by those of ordinary skill in the art without creative effort are within the scope of protection of the present invention.

[0084] It should be noted that all directional indications (such as up, down, left, right, front, back, etc.) in the embodiments of the present invention are only used to explain the relative positional relationship and movement of each component in a specific posture. If the specific posture changes, the directional indication will also change accordingly.

[0085] Furthermore, the use of terms such as "first" and "second" in this invention is for descriptive purposes only and should not be construed as indicating or implying their relative importance or implicitly specifying the number of technical features indicated. Therefore, features defined with "first" and "second" may explicitly or implicitly include at least one of those features. Additionally, the technical solutions of the various embodiments can be combined with each other, but only on the basis of being achievable by those skilled in the art. When the combination of technical solutions is contradictory or impossible to implement, such a combination of technical solutions should be considered non-existent and not within the scope of protection claimed by this invention.

[0086] Example 1

[0087] This embodiment provides a data center management method, such as Figure 1 As shown, it includes the following steps:

[0088] S1: Obtain power data of data center servers and calculate server energy consumption;

[0089] S2, obtain cooling data from the data center servers;

[0090] S3, establish the objective function based on cooling data and server energy consumption;

[0091] S4, obtains real-time data center server utilization and workload;

[0092] S5 uses the particle swarm optimization algorithm to solve the objective function;

[0093] S6. Based on the solution of the objective function, allocate tasks to the data center;

[0094] The power data includes: processor power data, memory power data, disk power data, and interface power data.

[0095] It should be noted that a data center contains multiple servers, and energy management of the data center can be broken down into managing each server individually. The parameters of each server's components differ, resulting in varying energy consumption for each server. Therefore, it is necessary to obtain server power data to measure server energy consumption. Data center server power data includes the power data of the processor, memory, disk, and interfaces. Server energy consumption is estimated using this power data. In addition, the data center provides cooling for each server to ensure it operates at a suitable temperature. Data center server utilization refers to the current utilization rate of the server's processor, and data center server workload refers to the size of tasks the server needs to process at that time; these two data points are used to measure the server's operating status. This embodiment considers both server energy consumption and the energy consumption required for cooling. Since the energy consumption required for cooling is considered comprehensively based on the server's workload and cooling conditions, an objective function is established and solved based on cooling data and server energy consumption, aiming to minimize total energy consumption. This yields a task allocation scheme that minimizes total energy consumption while considering server performance and workload.

[0096] In this embodiment, by modeling the server energy consumption based on the power data of the data center server, and considering the server's workload and cooling capacity, the total energy consumption of the server is modeled and solved to obtain a task allocation scheme with the lowest total energy consumption while ensuring sufficient cooling for each server, thereby reducing the energy consumption of the data center.

[0097] In some embodiments, step S1 involves acquiring power data of the data center server and calculating the server's energy consumption, such as... Figure 2 As shown, it includes the following steps:

[0098] S11, obtain the processor's full-load power and idle power;

[0099] S12, calculate the processor power consumption based on the processor's full-load power and idle power;

[0100] S13, obtain the memory read power, write power and refresh power;

[0101] S14, calculate memory energy consumption based on memory read power, memory write power and memory refresh power;

[0102] S15, obtain the disk's read power, write power, and idle power;

[0103] S16, Calculate disk energy consumption based on disk read power, disk write power and disk idle power;

[0104] S17, obtain the task size and bandwidth of the interface;

[0105] S18, Calculate the interface energy consumption based on the task size and bandwidth of the interface;

[0106] S19 calculates server energy consumption based on processor energy consumption, memory energy consumption, disk energy consumption, and interface energy consumption.

[0107] In some embodiments, the formula for calculating server energy consumption is as follows:

[0108] ,

[0109] ,

[0110] ,

[0111] ,

[0112] ,

[0113] in, For server power consumption, As the first weight, For processor power consumption, As the second weight, For memory power consumption, As the third weight, For disk power consumption, As the fourth weight, For interface power consumption, This represents the processor's full load power. The processor's idle power, For processor utilization, For memory read power, The power consumption for memory writes. For memory refresh power, This refers to the disk's read power. This refers to the disk's write power. This refers to the idle power of the disk. The first performance factor for the interface. The task size of the interface. The bandwidth of the interface. This is the second performance factor for the interface.

[0114] It should be noted that full-load power refers to the maximum power of this component, idle power refers to the power of this component when it is idle, and processor utilization refers to the processor's load percentage. Memory power consumption is calculated based on memory read, write, and refresh power, and disk power consumption is calculated based on disk read, write, and idle power, also reflecting their utilization. The first and second performance coefficients for the interfaces vary depending on the interface and can be considered constants. Since the energy consumption calculations for the four components require different times, but these times can be obtained based on the allocated workload, setting weighting coefficients for the four energy consumption models is beneficial for calculating the server's total energy consumption.

[0115] In some embodiments, step S3 involves establishing an objective function based on cooling data and server energy consumption, such as... Figure 3 As shown, it includes the following steps:

[0116] S31, Calculate the cooling efficiency of the data center servers based on the cooling data;

[0117] S32 calculates the total energy consumption of the data center servers based on cooling efficiency and server energy consumption;

[0118] S33. Establish an objective function based on minimizing the total energy consumption of data center servers.

[0119] In some embodiments, the total energy consumption of a data center server is calculated using the following formula:

[0120] ,

[0121] ,

[0122] in, This represents the total energy consumption of the data center servers. For server power consumption, For cooling energy consumption, For runtime, For cooling efficiency;

[0123] The objective function is expressed as follows:

[0124] ,

[0125] ,

[0126] ,

[0127] ,

[0128] in, Let be the objective function. For the first Total energy consumption of each server For the number of servers, For the first The utilization rate of each server For the first Cooling energy consumption of a single server For cooling threshold, For the first Task size per server This refers to the workload.

[0129] It should be noted that cooling data refers to the cooling temperature of the data center servers. Cooling efficiency and runtime can be calculated empirically. Based on this, the total energy consumption of the servers can be obtained from the server energy consumption and cooling energy consumption. An objective function is constructed by summing the total energy consumption of each server and minimizing it. The objective function is subject to three constraints: first, the utilization rate of each server must not exceed 1, meaning the server has sufficient capacity for task processing; second, cooling cannot exceed the cooling module's capacity limit, meaning cooling energy consumption is less than or equal to the cooling threshold; and third, the sum of tasks assigned to each server must equal the number of tasks to be assigned.

[0130] In some embodiments, step S5 involves solving the objective function using a particle swarm optimization algorithm, such as... Figure 4 As shown, it includes the following steps:

[0131] S51, randomly initialize particles to obtain the initial population;

[0132] S52, calculates the fitness value of the particle;

[0133] S53, based on the fitness values ​​of the particles, update the optimal fitness values ​​of the individual and the optimal fitness values ​​of the population;

[0134] S54, update the particle's velocity and position based on the individual's optimal fitness value and the group's optimal fitness value;

[0135] S55, repeatedly update the particle's velocity and position until the maximum number of iterations is reached;

[0136] S56, output the solution to the objective function to obtain the task allocation amount for the data center server;

[0137] The formula for calculating fitness value is as follows:

[0138] ,

[0139] in, For fitness value, For the first Total energy consumption of each server This refers to the number of servers.

[0140] It should be noted that by obtaining server utilization and workload, the energy consumption status of the data center at that moment is determined. At this point, only the task allocation of each server needs to be determined, and the total energy consumption of the servers can be calculated using the data center server total energy consumption calculation formula. Therefore, the position of the population particles is a vector composed of the task allocation of each server. By continuously optimizing the population particles, the optimal task allocation scheme is obtained. The fitness value is the reciprocal of the logarithm of the total energy consumption of the servers. Minimizing the objective function requires maximizing the fitness value. When updating the optimal fitness value, the larger of the two fitness values ​​is used.

[0141] In some embodiments, step S54 involves updating the particle's velocity and position based on the individual's optimal fitness value and the group's optimal fitness value, such as... Figure 5 As shown, it includes the following steps:

[0142] S541, the particle speed is updated based on the optimal fitness value of the individual and the optimal fitness value of the group to obtain the first update speed;

[0143] S542, update the particle position according to the first update speed to obtain the first updated position;

[0144] S543, Calculate the update fitness value of the particle based on the first update position;

[0145] S544, the difference between the updated fitness value and the individual's optimal fitness value is calculated to obtain the fitness difference value;

[0146] S545, when the fitness difference is greater than or equal to the preset difference, the first update speed is used as the particle speed and the first update position is used as the particle position;

[0147] S546, when the fitness difference is less than the preset difference, the mutation algorithm is used to update the first update speed and the first update position to obtain the second update speed and the second update position. The second update speed is used as the particle speed and the second update position is used as the particle position.

[0148] It should be noted that, considering the initial insufficient diversity of the particle population, if premature convergence is detected in the updated particle population, a mutation algorithm is used to perturb the particle population to prevent the algorithm from entering a local optimum. Specifically, the updates of particle velocity and position depend on the fitness value. When the change in fitness value is small, the algorithm may enter a local optimum. Therefore, after updating the particle velocity and position, the difference between the updated fitness value and the individual's optimal fitness value is calculated. When the difference is large, it indicates normal convergence, and optimization is performed using the updated position and update velocity. When the difference is less than a preset difference, the mutation algorithm is used to further modify the particle's position and velocity based on the updated position and update velocity.

[0149] The formula for the mutation algorithm is expressed as follows:

[0150] ,

[0151] in, For the second update speed, The coefficient of variation is 1. For the fastest update speed.

[0152] In this embodiment, the objective function is solved using the particle swarm optimization algorithm. When the problem gets stuck in a local optimum, the position and velocity of the particles are updated using a mutation algorithm. This improves the solution speed while ensuring the quality of the solution, resulting in the optimal task allocation scheme and reducing the energy consumption of the data center.

[0153] Example 2

[0154] This embodiment provides a data center management system, applying the data center management method described above, such as... Figure 6 As shown, it includes: server energy consumption calculation module, data center cooling module, objective function establishment module, real-time data monitoring module, objective function solving module, and task allocation module;

[0155] The server energy consumption calculation module is used to obtain power data of data center servers and calculate server energy consumption.

[0156] The data center cooling module is used to acquire cooling data from data center servers.

[0157] The objective function creation module is used to create an objective function based on cooling data and server energy consumption.

[0158] The real-time data monitoring module is used to obtain the utilization rate and workload of data center servers in real time.

[0159] The objective function solution module is used to solve the objective function using the particle swarm optimization algorithm;

[0160] The task allocation module is used to allocate tasks to the data center based on the solution of the objective function.

[0161] In this embodiment, by modeling the server energy consumption based on the power data of the data center server, and considering the server's workload and cooling capacity, the total energy consumption of the server is modeled and solved to obtain a task allocation scheme with the lowest total energy consumption while ensuring sufficient cooling for each server, thereby reducing the energy consumption of the data center.

[0162] It should be understood that the disclosed system can be implemented in other ways. For example, the system embodiments described above are merely illustrative. For instance, the module division described above is only a logical functional division; in actual implementation, there may be other division methods. For example, multiple modules or components may be combined or integrated into another system, or some features may be ignored or not executed. Furthermore, each functional module can be integrated into a processing module, or each module can exist physically separately, or two or more modules can be integrated into one module. The integrated modules described above can be implemented in hardware or as software functional modules.

[0163] Example 3

[0164] This embodiment provides an electronic device 2, such as... Figure 7 As shown, there is a processor 21 and a memory 22. The memory 22 is used to store computer program code, which includes computer instructions. When the processor 21 executes the computer instructions, the electronic device performs the aforementioned data center management method.

[0165] The electronic device 2 includes a processor 21, a memory 22, an output device 23, and an input device 24. The processor 21, memory 22, output device 23, and input device 24 are coupled together via connectors, which may include various interfaces, transmission lines, or buses, etc., and are not limited in this embodiment of the invention. It should be understood that in various embodiments of the invention, coupling refers to mutual connection through a specific method, including direct connection or indirect connection through other devices, such as through various interfaces, transmission lines, buses, etc.

[0166] The processor 21 can be one or more graphics processing units (GPUs). If the processor 21 is a GPU, the GPU can be a single-core GPU or a multi-core GPU. Optionally, the processor 21 can be a processor group composed of multiple GPUs, with the multiple processors coupled to each other via one or more buses. Optionally, the processor 21 can also be other types of processors, etc., and this embodiment of the invention is not limited thereto.

[0167] The memory 22 can be used to store computer program instructions, as well as various types of computer program code, including program code for executing the present invention. Optionally, the memory 22 may include, but is not limited to, random access memory (RAM), read-only memory (ROM), erasable programmable read-only memory (EPROM), or compact disc read-only memory (CD-ROM), and the memory 22 is used for related instructions and data.

[0168] Input device 24 is used to input data and / or signals, and output device 23 is used to output data and / or signals. Output device 23 and input device 24 can be independent devices or an integrated device.

[0169] This embodiment provides a computer-readable storage medium storing a computer program, which includes program instructions. When executed by a processor of an electronic device, the program instructions cause the processor to perform the aforementioned data center management method.

[0170] The above description is merely a specific embodiment of the present invention, enabling those skilled in the art to understand or implement the invention. Various modifications to these embodiments will be readily apparent to those skilled in the art, and the general principles defined herein may be implemented in other embodiments without departing from the spirit or scope of the invention. Therefore, the invention is not to be limited to the embodiments shown herein, but is to be accorded the widest scope consistent with the principles and novel features claimed herein.

Claims

1. A data center management method, characterized in that, Includes the following steps: Obtain power data from data center servers and calculate server energy consumption; Obtain the cooling data of the data center server; Based on the cooling data and the server energy consumption, establish an objective function; Real-time acquisition of the utilization rate and workload of the data center servers; The objective function is solved using the particle swarm optimization algorithm. Based on the solution of the objective function, tasks are allocated to the data center; The power data includes: processor power data, memory power data, disk power data, and interface power data; The formula for calculating server energy consumption is as follows: ； in, For server power consumption, As the first weight, For processor power consumption, As the second weight, For memory power consumption, As the third weight, For disk power consumption, As the fourth weight, For interface power consumption, This represents the processor's full load power. The processor's idle power, For processor utilization, For memory read power, The power consumption for memory writes. For memory refresh power, This refers to the disk's read power. This refers to the disk's write power. This refers to the idle power of the disk. The first performance factor for the interface. The task size of the interface. The bandwidth of the interface. This is the second performance coefficient for the interface; The step of establishing an objective function based on the cooling data and the server energy consumption includes the following steps: Calculate the cooling efficiency of the data center server based on the cooling data. Calculate the total energy consumption of the data center server based on the cooling efficiency and the server energy consumption; The objective function is established based on minimizing the total energy consumption of the data center servers. The formula for calculating the total energy consumption of the data center server is as follows: ； in, This represents the total energy consumption of the data center servers. For server power consumption, For cooling energy consumption, For runtime, For cooling efficiency; The objective function is expressed as follows: ； in, Let be the objective function. For the first Total energy consumption of each server For the number of servers, For the first The utilization rate of each server For the first Cooling energy consumption of a single server For cooling threshold, For the first Task size per server For task volume; The objective function is solved using the particle swarm optimization algorithm, including the following steps: Randomly initialize the particles to obtain the initial population; Calculate the fitness value of the particles; Update the optimal fitness value of the individual and the optimal fitness value of the population based on the fitness value of the particles. The velocity and position of the particle are updated based on the optimal fitness value of the individual and the optimal fitness value of the population. The velocity and position of the particle are repeatedly updated until the maximum number of iterations is reached. Output the solution to the objective function to obtain the task allocation amount for the data center server; The formula for calculating the fitness value is as follows: ； in, For fitness value, For the first Total energy consumption of each server The number of servers; The step of updating the velocity and position of the particle based on the optimal fitness value of the individual and the optimal fitness value of the population includes the following steps: The velocity of the particle is updated based on the optimal fitness value of the individual and the optimal fitness value of the population to obtain the first update velocity; The position of the particle is updated according to the first update rate to obtain the first updated position; Calculate the update fitness value of the particle based on the first update position; The fitness difference is obtained by subtracting the updated fitness value from the individual's optimal fitness value. When the fitness difference is greater than or equal to a preset difference, the first update speed is taken as the speed of the particle, and the first update position is taken as the position of the particle. When the fitness difference is less than the preset difference, the first update speed and the first update position are updated using a mutation algorithm to obtain a second update speed and a second update position. The second update speed is used as the speed of the particle, and the second update position is used as the position of the particle.

2. The data center management method according to claim 1, characterized in that, The process of obtaining power data from data center servers and calculating server energy consumption includes the following steps: Obtain the processor's full-load power and idle power; Calculate the processor's energy consumption based on the processor's full-load power and idle power. Obtain the memory read power, write power, and refresh power; Calculate the memory power consumption based on the memory read power, the memory write power, and the memory refresh power; Obtain the disk's read power, write power, and idle power; Calculate disk energy consumption based on the disk's read power, write power, and idle power. Get the task size and bandwidth of the interface; Calculate the interface power consumption based on the task size and bandwidth of the interface; The server energy consumption is calculated based on the processor energy consumption, the memory energy consumption, the disk energy consumption, and the interface energy consumption.

3. A data center management system, employing the data center management method as described in any one of claims 1 to 2, characterized in that, include: The module includes a server energy consumption calculation module, a data center cooling module, an objective function establishment module, a real-time data monitoring module, an objective function solving module, and a task allocation module. The server energy consumption calculation module is used to obtain power data of the data center server and calculate the server energy consumption. The data center cooling module is used to acquire cooling data from the data center servers; The objective function establishment module is used to establish an objective function based on the cooling data and the server energy consumption; The real-time data monitoring module is used to obtain the utilization rate and workload of the data center server in real time. The objective function solving module is used to solve the objective function using a particle swarm optimization algorithm; The task allocation module is used to allocate tasks to the data center based on the solution of the objective function.

4. An electronic device, characterized in that, The device includes a processor and a memory, the memory being used to store computer program code, the computer program code including computer instructions, wherein when the processor executes the computer instructions, the electronic device performs a data center management method as described in any one of claims 1 to 2.

5. A computer-readable storage medium, characterized in that, The computer-readable storage medium stores a computer program, the computer program including program instructions, which, when executed by a processor of an electronic device, cause the processor to perform a data center management method as described in any one of claims 1 to 2.

Images