CN-120103958-B - Data center management method, system and storage medium

CN120103958BCN 120103958 BCN120103958 BCN 120103958BCN-120103958-B

Abstract

The invention discloses a data center management method, a system and a storage medium, and relates to the technical field of data processing, wherein the method comprises the steps of obtaining power data of a data center server and calculating energy consumption of the server; the method comprises the steps of obtaining cooling data of a data center server, establishing an objective function according to the cooling data and the energy consumption of the server, obtaining the utilization rate and the task quantity of the data center server in real time, solving the objective function by adopting a particle swarm algorithm, and distributing tasks to the data center according to the solution of the objective function, wherein the power data comprise processor power data, memory power data, disk power data and interface power data. According to the invention, the energy consumption of the server is modeled according to the power data of the data center server, the task quantity and the cooling capacity of the server are considered, the total energy consumption of the server is modeled and solved, the task allocation scheme with the lowest total energy consumption and sufficient cooling of each server is ensured, and the energy consumption of the data center is reduced.

Inventors

PAN YONGJIAN
DU MIN
CHEN DONGDONG
LUO HUASHENG
Weng Jianrui

Assignees

广州宽恒信息科技有限公司

Dates

Publication Date: 20260508
Application Date: 20250227

Claims (5)

1. A data center management method, comprising the steps of: acquiring power data of a data center server, and calculating energy consumption of the server; Acquiring cooling data of the data center server; Establishing an objective function according to the cooling data and the server energy consumption; acquiring the utilization rate and the task quantity of the data center server in real time; Solving the objective function by adopting a particle swarm algorithm; Task allocation is carried out on the data center according to the solution of the objective function; The power data comprises processor power data, memory power data, disk power data and interface power data; the calculation formula of the server energy consumption is as follows: ; Wherein, the In order to achieve the energy consumption of the server, As a first weight to be used, In order for the processor to consume power, As a result of the second weight being set, For the purpose of energy consumption of the internal memory, As a result of the third weight being given, For the purpose of energy consumption of the magnetic disk, For the fourth weight to be the fourth weight, In order to achieve the energy consumption of the interface, For the full power of the processor, For the idle power of the processor, In order for the processor to be utilized, For the read power of the memory, Is the write power of the memory and, For the refresh power of the memory, For the read power of the disk, For the write power of the disk, For the idle power of the disk, As a first coefficient of performance of the interface, For the task size of the interface, For the bandwidth size of the interface, A second coefficient of performance for the interface; The objective function is established according to the cooling data and the energy consumption of the server, and the method comprises the following steps: According to the cooling data, calculating cooling efficiency of the data center server; calculating the total energy consumption of the data center server according to the cooling efficiency and the energy consumption of the server; establishing the objective function according to the total energy consumption minimization of the data center server as an objective; The total energy consumption of the data center server is calculated as follows: ; Wherein, the The total energy consumption of the data center server, In order to achieve the energy consumption of the server, In order to supply the energy consumption for cooling, In order for the run-time to be run, Is the cooling efficiency; The formula of the objective function is as follows: ; Wherein, the As a function of the object to be processed, Is the first The total energy consumption of the individual servers is, For the number of servers to be used, Is the first The utilization rate of the individual servers is determined, Is the first The cooling energy consumption of the individual servers, For the purpose of the cooling threshold value, Is the first The task size of the individual servers, Is the task quantity; Solving the objective function by adopting a particle swarm algorithm, comprising the following steps: Randomly initializing particles to obtain an initial population; Calculating the fitness value of the particles; updating the optimal fitness value of the individual and the optimal fitness value of the population according to the fitness value of the particle; Updating the speed of the particles and the positions of the particles according to the optimal fitness value of the individual and the optimal fitness value of the population; repeatedly updating the speed of the particles and the positions of the particles until the iteration number reaches the maximum iteration number; outputting the solution of the objective function to obtain the task allocation amount of the data center server; The calculation formula of the fitness value is as follows: ; Wherein, the In order to adapt the value of the degree of adaptation, Is the first The total energy consumption of the individual servers is, Is the number of servers; said updating the speed and position of said particles according to the optimal fitness value of said individual and the optimal fitness value of said population, comprising the steps of: Updating the speed of the particles according to the optimal fitness value of the individual and the optimal fitness value of the population to obtain a first updating speed; updating the position of the particle according to the first updating speed to obtain a first updating position; calculating an updated fitness value of the particle according to the first updated position; performing difference between the updated fitness value and the optimal fitness value of the individual to obtain a fitness difference value; when the fitness difference value is larger than or equal to a preset difference value, taking the first updating speed as the speed of the particles and taking the first updating position as the position of the particles; and when the adaptability difference value is smaller than the preset difference value, updating the first updating speed and the first updating position by adopting a mutation algorithm to obtain a second updating speed and a second updating position, wherein the second updating speed is used as the speed of the particles, and the second updating position is used as the position of the particles.
2. The data center management method according to claim 1, wherein the steps of obtaining power data of the data center server, calculating server power consumption, and comprising: acquiring full load power and no-load power of a processor; calculating the energy consumption of the processor according to the full load power of the processor and the idle load power of the processor; Acquiring the read power, the write power and the refresh power of the memory; calculating memory energy consumption according to the read power of the memory, the write power of the memory and the refresh power of the memory; obtaining the read power, the write power and the idle power of a magnetic disk; Calculating the energy consumption of the magnetic disk according to the read power of the magnetic disk, the write power of the magnetic disk and the idle power of the magnetic disk; Acquiring the task size and the bandwidth size of an interface; calculating interface energy consumption according to the task size of the interface and the bandwidth size of the interface; and calculating the energy consumption of the server according to the energy consumption of the processor, the energy consumption of the memory, the energy consumption of the disk and the energy consumption of the interface.
3. A data center management system, applying the data center management method according to any one of claims 1 to 2, comprising a server energy consumption calculation module, a data center cooling module, an objective function establishment module, a data real-time monitoring module, an objective function solving module and a task allocation module; the server energy consumption calculation module is used for acquiring power data of the data center server and calculating the energy consumption of the server; the data center cooling module is used for acquiring cooling data of the data center server; the objective function building module is used for building an objective function according to the cooling data and the server energy consumption; the data real-time monitoring module is used for acquiring the utilization rate and the task quantity of the data center server in real time; The objective function solving module is used for solving the objective function by adopting a particle swarm algorithm; and the task allocation module is used for allocating tasks to the data center according to the solution of the objective function.
4. An electronic device comprising a processor and a memory for storing computer program code, the computer program code comprising computer instructions which, when executed by the processor, perform a data center management method as claimed in any one of claims 1 to 2.
5. A computer readable storage medium, characterized in that the computer readable storage medium has stored therein a computer program comprising program instructions which, when executed by a processor of an electronic device, cause the processor to perform a data center management method according to any of claims 1 to 2.

Description

Data center management method, system and storage medium Technical Field The present invention relates to the field of data processing technologies, and in particular, to a data center management method, a system, and a storage medium. Background With the rapid development of cloud computing, data centers have become an important part of the running information society. However, the data center has high power and consumes a large amount of energy in the operation process, so that the data center has the problems of high operation cost and low system reliability, and therefore, the data center needs to be managed, and the energy consumption of the data center is reduced. Existing data center energy consumption reduction schemes optimize data centers by managing the data centers on a single level, such as reducing the deployment of the number of servers, and performing tasks with a small number of servers. However, a large number of tasks are piled up in a small number of servers, the load of the servers is large, the computing performance of the system is reduced due to long-time work, the tasks are slow to process, and high energy consumption still exists in the data center. Disclosure of Invention The invention aims to provide a data center management method, a system and a storage medium, which model the energy consumption of a server according to the power data of the data center server, consider the task quantity and the cooling capacity of the server, modeling and solving are carried out on the total energy consumption of the servers, so that a task allocation scheme which is minimum in total energy consumption and ensures sufficient cooling of each server is obtained, and the energy consumption of the data center is reduced. The aim of the invention is realized by adopting the following technical modes: in a first aspect, the present invention provides a data center management method, including the steps of: acquiring power data of a data center server, and calculating energy consumption of the server; Acquiring cooling data of the data center server; Establishing an objective function according to the cooling data and the server energy consumption; acquiring the utilization rate and the task quantity of the data center server in real time; Solving the objective function by adopting a particle swarm algorithm; Task allocation is carried out on the data center according to the solution of the objective function; The power data comprises processor power data, memory power data, disk power data and interface power data. Preferably, the step of obtaining the power data of the data center server and calculating the energy consumption of the server includes the following steps: acquiring full load power and no-load power of a processor; calculating the energy consumption of the processor according to the full load power of the processor and the idle load power of the processor; Acquiring the read power, the write power and the refresh power of the memory; calculating memory energy consumption according to the read power of the memory, the write power of the memory and the refresh power of the memory; obtaining the read power, the write power and the idle power of a magnetic disk; Calculating the energy consumption of the magnetic disk according to the read power of the magnetic disk, the write power of the magnetic disk and the idle power of the magnetic disk; Acquiring the task size and the bandwidth size of an interface; calculating interface energy consumption according to the task size of the interface and the bandwidth size of the interface; and calculating the energy consumption of the server according to the energy consumption of the processor, the energy consumption of the memory, the energy consumption of the disk and the energy consumption of the interface. Preferably, the calculation formula of the server energy consumption is as follows: , , , , , Wherein, the In order to achieve the energy consumption of the server,As a first weight to be used,In order for the processor to consume power,As a result of the second weight being set,For the purpose of energy consumption of the internal memory,As a result of the third weight being given,For the purpose of energy consumption of the magnetic disk,For the fourth weight to be the fourth weight,In order to achieve the energy consumption of the interface,For the full power of the processor,For the idle power of the processor,In order for the processor to be utilized,For the read power of the memory,Is the write power of the memory and,For the refresh power of the memory,For the read power of the disk,For the write power of the disk,For the idle power of the disk,As a first coefficient of performance of the interface,For the task size of the interface,For the bandwidth size of the interface,Is the second coefficient of performance of the interface. Preferably, the establishing an objective function according to the cooling data and the server energy consu