A method and system for reducing the PUE value of an intelligent computing center

By acquiring low-load servers in the intelligent computing center and calculating task migration priority ranking based on their number, and determining the priority sequence based on computing power factor and energy consumption utilization, the problem of unreasonable allocation of calculation tasks is solved and the server load is realized Increase and energy consumption reduce further reduce the PUE value.

CN118295798BActive Publication Date: 2024-11-26SHANGRAO GAOTOU ZHICHENG TECH CO LTD
View PDF 0 Cites 1 Cited by

Patent Information

Application Number
CN202410370989.9
Authority / Receiving Office
CN · China
Patent Type
Patents(China)
Current Assignee / Owner
Filing Date
2024-03-29
Publication Date
2024-11-26
Estimated Expiration
2044-03-29

AI Technical Summary

Technical Problem

The existing technology has unreasonable allocation of calculation tasks when reducing the PUE value of the intelligent computing center, resulting in low server load improvement efficiency and insufficient energy-consuming equipment resources.

Method used

By acquiring low-load servers and calculating task migration priority rankings based on their number, until the server is in full load state, the priority sequence of the computer room is determined based on the computing power factor, low load proportion and energy consumption utilization rate, and the calculation tasks are reasonably allocated. And adjust the refrigeration system after task assignment to reduce energy consumption.

Benefits of technology

The rational allocation of computing tasks is achieved, the server load and overall energy consumption is improved, the PUE value is reduced, and the energy consumption is further reduced by adjusting the power of the energy-consuming equipment.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN118295798B_ABST
    Figure CN118295798B_ABST
Patent Text Reader

Abstract

The present invention discloses a method and system for reducing the PUE value of an intelligent computing center, which is used to reduce the PUE value of an intelligent computing center. The intelligent computing center includes multiple computer rooms for arranging servers, and multiple server arrangement areas arranged side by side are arranged in the computer rooms. The servers are distributed in the server arrangement areas. The method includes: obtaining a target server in a low-load operation state from the server, and determining the server arrangement area to which the target server belongs; sorting the computing task migration priority of the server arrangement area according to the number of target servers in the server arrangement area to determine the computing task migration priority of the server arrangement area; obtaining several computing tasks outside the intelligent computing center, and migrating the computing tasks to the target servers in the current server arrangement area according to the computing task migration priority, until all the target servers in the current server arrangement area are in a full-load operation state. The present invention solves the problem of unreasonable computing task allocation in the prior art when reducing the PUE value.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] The present invention relates to the field of data processing technology, and in particular to a method and system for reducing the PUE value of an intelligent computing center. Background Art

[0002] PUE (Power Usage Effectiveness) is the ratio of all energy consumed by the intelligent computing (data) center to the energy used by the IT load. The PUE value is an important indicator for evaluating the energy efficiency of data centers. At present, the actual PUE value of most data centers deviates greatly from the theoretical PUE value, which is mainly caused by insufficient load on the data center servers. For example, most of the servers in the data center are in a low-load operating state.

[0003] Therefore, in order to reduce the PUE value of the data center, most of the computing tasks outside the data center are migrated to the servers in the data center for computing, thereby increasing the load of the server, and then increasing the energy consumption of the server, and finally achieving the purpose of reducing the PUE value of the data center. However, in the prior art, when migrating and allocating computing tasks, the computing tasks are randomly sent to the server. When the computing tasks outside the data center cannot increase all low-load servers in the data center to high loads, the random allocation method is not reasonable enough. Summary of the invention

[0004] In view of this, the purpose of the present invention is to provide a method and system for reducing the PUE value of an intelligent computing center, aiming to solve the problem of unreasonable allocation of computing tasks when reducing the PUE value in the prior art.

[0005] The present invention is achieved in that:

[0006] A method for reducing the PUE value of an intelligent computing center is used to reduce the PUE value of the intelligent computing center, wherein the intelligent computing center includes a plurality of computer rooms for arranging servers, wherein the computer rooms are provided with a plurality of server arrangement areas arranged side by side, and the servers are distributed in the server arrangement areas, and the method includes:

[0007] Acquire a target server in a low-load operation state from the servers, and determine a server deployment area to which the target server belongs;

[0008] According to the number of the target servers in the server deployment area, the server deployment area is prioritized for computing task migration to determine the computing task migration priority of the server deployment area;

[0009] Acquire several computing tasks outside the intelligent computing center, and migrate the computing tasks to the target servers in the current server deployment area according to the computing task migration priority until all the target servers in the current server deployment area are in full load operation state.

[0010] Furthermore, the method for reducing the PUE value of the intelligent computing center, wherein the step of obtaining a plurality of computing tasks outside the intelligent computing center and migrating the computing tasks to the target servers in the current server deployment area according to the computing task migration priority until all the target servers in the current server deployment area are in full load operation state, further includes:

[0011] Obtaining the remaining computing power of each server in each of the computer rooms, so as to determine the computing power factor of the computer room according to the remaining computing power of each server;

[0012] Determine the low load ratio of the computer room according to the target server and the total number of servers in the computer room that are under low load, and determine the energy utilization rate of the computer room according to the percentage ratio of the server energy consumption of the computer room to the energy consumption of the computer room;

[0013] Determine the priority sequence of the computer room according to the computing power factor, the low load ratio and the energy consumption utilization rate according to preset rules;

[0014] The step of obtaining a plurality of computing tasks outside the intelligent computing center and migrating the computing tasks to the target servers in the current server deployment area according to the computing task migration priority until all the target servers in the current server deployment area are in a full load operation state comprises:

[0015] When task migration is required, the target computer room that currently needs to perform task migration is determined according to the priority sequence of the computer room, and the current server layout area is determined from the target computer room according to the computing task migration priority, and the computing task is migrated to the target server in the current server layout area.

[0016] Furthermore, in the above method for reducing the PUE value of the intelligent computing center, the step of determining the priority sequence of the computer room according to the computing power factor, the low load ratio and the energy consumption utilization rate according to a preset rule includes:

[0017] A preset coordinate system is established with low load proportion and energy consumption utilization rate as the horizontal and vertical coordinates of the coordinate system respectively;

[0018] Mapping the computer room to the preset coordinate system according to the low load proportion and energy utilization rate of the computer room;

[0019] Taking the coordinate origin as a base point, a plurality of sector-shaped areas with a preset radius are divided to determine a plurality of priority determination areas, and the basic priority of the computer room is determined according to the priority determination area where the computer room is located.

[0020] Further, the method for reducing the PUE value of the intelligent computing center, wherein the step of taking the coordinate origin as the base point and dividing a plurality of sector-shaped areas with a preset radius to determine a plurality of priority determination areas, and determining the basic priority of the computer room according to the priority determination area in which the computer room is located, further includes:

[0021] Obtain the computing power factor, low load ratio and energy utilization rate of the computer room within the basic priority level respectively, and determine the ranking factor of the computer room according to the computing power factor and low load ratio;

[0022] The sub-priority of the computer room is determined according to the energy consumption utilization rate and the ranking factor, and the priority sequence of the computer room is determined according to the basic priority and the sub-priority.

[0023] Furthermore, in the above method for reducing the PUE value of the intelligent computing center, the calculation formula for determining the secondary priority of the computer room according to the energy consumption utilization rate and the ranking factor is:

[0024]

[0025] Among them, μ j is the energy utilization rate of the jth computer room, ε i is the adjustment coefficient of the ith ranking factor, L ij is the value of the i-th sorting factor of the j-th computer room, N is a positive integer, 0≤N≤2.

[0026] Further, in the above method for reducing the PUE value of the intelligent computing center, each server layout area is provided with a separate cooling system, and the method further includes:

[0027] After the computing task allocation is completed, a target server deployment area in which the number of target servers is greater than a preset number is determined, and a refrigeration system of the target server deployment area is adjusted to reduce energy consumption of the refrigeration system.

[0028] Furthermore, in the above method for reducing the PUE value of the intelligent computing center, before the step of adjusting the refrigeration system of the target server layout area to reduce the energy consumption of the refrigeration system, the step further includes:

[0029] Determining whether the target server layout area meets the refrigeration system adjustment conditions;

[0030] If so, the step of adjusting the refrigeration system of the target server deployment area to reduce the energy consumption of the refrigeration system is performed.

[0031] Another object of the present invention is to provide a system for reducing the PUE value of an intelligent computing center, which is used to reduce the PUE value of the intelligent computing center. The intelligent computing center includes a plurality of computer rooms for arranging servers, and a plurality of server arrangement areas arranged side by side are arranged in the computer rooms. The servers are distributed in the server arrangement areas. The system includes:

[0032] An acquisition module, used for acquiring a target server in a low-load operation state from the servers, and determining a server deployment area to which the target server belongs;

[0033] A sorting module, configured to sort the computing task migration priorities of the server deployment area according to the number of the target servers in the server deployment area, so as to determine the computing task migration priorities of the server deployment area;

[0034] A migration module is used to obtain several computing tasks outside the intelligent computing center, and migrate the computing tasks to the target servers in the current server deployment area according to the computing task migration priority until all the target servers in the current server deployment area are in full load operation state.

[0035] Another object of the present invention is to provide a readable storage medium having a computer program stored thereon, wherein the program, when executed by a processor, implements the steps of any of the methods described above.

[0036] Another object of the present invention is to provide an electronic device, comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor implements the steps of any of the above-described methods when executing the program.

[0037] The present invention obtains a target server in a low-load operating state from the servers, and determines the server layout area to which the target server belongs; sorts the computing task migration priorities of the server layout area according to the number of target servers in the server layout area to determine the computing task migration priority of the server layout area; obtains computing tasks outside a number of intelligent computing centers, and migrates the computing tasks to the target servers in the current server layout area according to the computing task migration priority, that is, according to the established rules, the tasks of the servers in the server layout area are migrated in sequence, and the tasks of the servers in the server layout area with the front ranking are migrated in an orderly manner first, thereby realizing a reasonable allocation of computing task migration, and solving the problem of unreasonable computing task allocation in the prior art due to the reduction of PUE value efficiency. In addition, since the load of the servers in each server layout area is increased first and then the load of the servers in other server layout areas is increased, the possibility of adjusting the energy-consuming equipment in the server layout area with a large number of target servers is retained, and the energy consumption is reduced by adjusting the power of the energy-consuming equipment, thereby further reducing the PUE value. BRIEF DESCRIPTION OF THE DRAWINGS

[0038] Figure 1 This is a flow chart of a method for reducing the PUE value of an intelligent computing center in the first embodiment of the present invention;

[0039] Figure 2 A schematic diagram of determining the basic priority of a computer room in a method for reducing the PUE value of an intelligent computing center in a first embodiment of the present invention;

[0040] Figure 3 This is a structural block diagram of a system for reducing the PUE value of an intelligent computing center in the third embodiment of the present invention.

[0041] The following specific implementation manner will further illustrate the present invention in conjunction with the above-mentioned drawings. DETAILED DESCRIPTION

[0042] In order to facilitate understanding of the present invention, the present invention will be described more fully below with reference to the relevant drawings. Several embodiments of the present invention are given in the drawings. However, the present invention can be implemented in many different forms and is not limited to the embodiments described herein. On the contrary, the purpose of providing these embodiments is to make the disclosure of the present invention more thorough and comprehensive.

[0043] It should be noted that when an element is referred to as being "fixed to" another element, it may be directly on the other element or there may be a central element. When an element is considered to be "connected to" another element, it may be directly connected to the other element or there may be a central element at the same time. The terms "vertical", "horizontal", "left", "right" and similar expressions used herein are for illustrative purposes only.

[0044] Unless otherwise defined, all technical and scientific terms used herein have the same meaning as those commonly understood by those skilled in the art of the present invention. The terms used herein in the specification of the present invention are only for the purpose of describing specific embodiments and are not intended to limit the present invention. The term "and / or" used herein includes any and all combinations of one or more of the related listed types.

[0045] The following will explain in detail how to solve the problem of unreasonable allocation of computing tasks when reducing the PUE value in conjunction with specific embodiments and drawings.

[0046] Embodiment 1

[0047] See also Figure 1 , shown is a method for reducing the PUE value of an intelligent computing center in a first embodiment of the present invention, which is used to reduce the PUE value of an intelligent computing center. The intelligent computing center includes a plurality of computer rooms for arranging servers. The computer rooms are provided with a plurality of server arrangement areas arranged side by side. The servers are distributed in the server arrangement areas. The method includes steps S10 to S12.

[0048] Step S10, obtaining a target server in a low-load operation state from the servers, and determining a server deployment area to which the target server belongs.

[0049] Among them, the intelligent computing center generally includes multiple computer rooms, and multiple servers are arranged in the computer rooms. Specifically, the computer rooms are generally arranged in regular shapes, such as regular rectangles, in order to place the servers regularly. More specifically, each computer room is provided with multiple server layout areas arranged side by side and spaced apart, and each server layout area is used to layout servers. In the specific implementation, each server layout area is provided with a rack, and the server is placed on the rack. In the embodiment of the present invention, a complete row of racks can be used as a server layout area, and the racks in the same straight line in the computer room can be used as a server layout area, for example, racks in the same column or row in the computer room can be used as a server layout area.

[0050] Specifically, whether it is a low-load target server is determined based on whether the server's own load rate is lower than a set threshold. For example, a server with a load lower than 85% of its own load capacity is defined as a low-load target server. On the contrary, if the server's load capacity is higher than 85% of its own load capacity, the server is defined as a full-load server. In addition, since the server deployment area is determined in advance, the server deployment area to which the target server belongs can be determined after determining the target server in a low-load operation state. In specific implementation, generally, each server is configured with a unique identifier, and the server deployment area corresponding to each server is stored in advance, so that the server deployment area to which the server belongs can be determined by the identifier information.

[0051] Step S11 : sorting the computing task migration priorities of the server deployment areas according to the number of the target servers in the server deployment areas to determine the computing task migration priorities of the server deployment areas.

[0052] Among them, after obtaining the status of each server, the number of low-load target servers in each server layout area can be clearly determined. Therefore, the server layout areas can be sorted according to the number of target servers, so that when subsequent task migration is performed, the tasks can be migrated in sequence according to the sorting. For example, the server layout area with a large number of target servers is set as the server layout area for priority task migration, so that the server layout area with a very low load can be used preferentially, or the server layout area with a small number of target servers is set as the server layout area for priority task migration, so that the server layout area in a fully loaded state can be obtained preferentially. In this embodiment, the server layout area with a small number of target servers is set as the server layout area for priority task migration.

[0053] Step S12, obtain several computing tasks outside the intelligent computing center, and migrate the computing tasks to the target servers in the current server deployment area according to the computing task migration priority until all the target servers in the current server deployment area are in full load operation state.

[0054] Among them, computing tasks outside the intelligent computing center are obtained, that is, computing tasks outside the intelligent computing center are processed by the intelligent computing center, and the remaining resources of the intelligent computing center are reasonably utilized. Moreover, the overall load of the servers in the intelligent computing center is improved through the computing tasks to improve the IT energy consumption of the intelligent computing center, thereby reducing the PUE value of the intelligent computing center. In the specific implementation, the computing tasks are preferentially migrated to the target servers in the server layout area at the front end of the sorting according to the prior priority sorting. When the computing tasks need to be migrated to another server layout area, all the target servers in the current server layout area need to be in full load operation state.

[0055] It can be understood that by migrating computing tasks after sorting the priorities, it is possible to reasonably and quickly obtain server layout areas where multiple servers are fully loaded. In practice, the server layout areas will be equipped with some other energy-consuming equipment that are essential for the intelligent computing center, such as refrigeration systems, etc., which generally operate at rated power, that is, the default server layout area runs at the fully loaded rated power under the current external environment, and quickly obtains server layout areas where multiple servers are fully loaded, avoiding excess resources of other energy-consuming equipment. On the contrary, the server layout area remaining after the task allocation is completed is the server layout area with a large number of target servers in the intelligent computing center. After completing the allocation of computing tasks, it is guaranteed that the server layout area with the largest number of target servers still exists, increasing the possibility of adjusting or shutting down other energy-consuming equipment in the server layout area, mainly shutting down, and adjusting these devices according to actual conditions, thereby reducing the energy consumption of these devices and further reducing the PUE value.

[0056] Specifically, in some optional embodiments of the present invention, each server arrangement area is correspondingly provided with a separate cooling system, and the method further includes:

[0057] After the computing task allocation is completed, a target server deployment area in which the number of target servers is greater than a preset number is determined, and a refrigeration system of the target server deployment area is adjusted to reduce energy consumption of the refrigeration system.

[0058] Among them, since server operations need to generate a large amount of heat, it is necessary to cool down the temperature in the computer room. Therefore, a refrigeration system is installed in the computer room. For example, an air conditioner is installed in the computer room. In the embodiment of the present invention, the air outlet of the air conditioner is arranged above each server layout area, so that the temperature in the area can be directly reduced, and the air outlet of the air conditioner above each server layout area can be controlled separately. In the embodiment of the present invention, the server layout area with a small number of target servers is set as the server layout area for priority task migration, that is, all servers in the server layout area can be quickly adjusted to a full load state. Correspondingly, after the computing task allocation is completed, those server layout areas that originally had a large number of target servers still maintain a large number of target servers. Since there are many servers with low load, the heat generated in the server layout area at this time is less. Therefore, the refrigeration system can be properly adjusted or even shut down to reduce the energy consumption of the refrigeration system, reduce other energy consumption of the data center, and further reduce the PUE value.

[0059] In addition, in some optional embodiments of the present invention, before the step of adjusting the refrigeration system of the target server arrangement area to reduce the energy consumption of the refrigeration system, the step further includes:

[0060] Determining whether the target server layout area meets the refrigeration system adjustment conditions;

[0061] If so, the step of adjusting the refrigeration system of the target server deployment area to reduce the energy consumption of the refrigeration system is performed.

[0062] Among them, before making an adjustment, it is determined whether the adjustment conditions are met, that is, whether the refrigeration system can be adjusted at present, so that the refrigeration system can be adjusted after it can be adjusted. In addition, in some optional embodiments of the present invention, it is also possible to determine whether the refrigeration system in the current target server layout area can be shut down, so that the refrigeration system in the target server layout area can be shut down.

[0063] In summary, the method for reducing the PUE value of the intelligent computing center in the above embodiment of the present invention obtains a target server in a low-load operating state from the server, and determines the server layout area to which the target server belongs; sorts the computing task migration priority of the server layout area according to the number of target servers in the server layout area to determine the computing task migration priority of the server layout area; obtains several computing tasks outside the intelligent computing center, and migrates the computing tasks to the target servers in the current server layout area according to the computing task migration priority, that is, according to the established rules, the tasks of the servers in the server layout area are migrated in sequence, and the tasks of the servers in the server layout area with the front ranking are migrated in an orderly manner, thereby realizing the reasonable allocation of computing task migration, and solving the problem of unreasonable computing task allocation when reducing the PUE value efficiency in the prior art. In addition, since the load of the servers in each server layout area is increased first and then the load of the servers in other server layout areas is increased, the possibility of adjusting the energy-consuming equipment in the server layout area with a large number of target servers is retained, and the energy consumption is reduced by adjusting the power of the energy-consuming equipment, thereby further reducing the PUE value.

[0064] Embodiment 2

[0065] This embodiment also proposes a method for reducing the PUE value of the intelligent computing center. The method for reducing the PUE value of the intelligent computing center proposed in this embodiment is different from the method for reducing the PUE value of the intelligent computing center proposed in the first embodiment in that:

[0066] Before step S12, the following steps are also included:

[0067] Obtaining the remaining computing power of each server in each of the computer rooms, so as to determine the computing power factor of the computer room according to the remaining computing power of each server;

[0068] Determine the low load ratio of the computer room according to the target server and the total number of servers in the computer room that are under low load, and determine the energy utilization rate of the computer room according to the percentage ratio of the server energy consumption of the computer room to the energy consumption of the computer room;

[0069] The priority sequence of the computer rooms is determined according to preset rules based on the computing power factor, low load ratio and energy utilization rate.

[0070] Step S12 also includes:

[0071] When task migration is required, the target computer room that currently needs to perform task migration is determined according to the priority sequence of the computer room, and the current server layout area is determined from the target computer room according to the computing task migration priority, and the computing task is migrated to the target server in the current server layout area.

[0072] Among them, since there are multiple computer rooms in the data center, before the priority sorting of the server layout area in the computer room is carried out, the computer rooms are also sorted once to first determine the computer room for migration task calculation. Specifically, the computing power factor of the computer room is a value of the remaining computing tasks that can be processed by the computer room, the low load proportion, that is, the proportion of low-load servers in the computer room to the total servers in the computer room, which represents the coverage of low-load servers in the computer room, and the current energy utilization rate of the computer room. The current energy utilization rate of the computer room refers to the ratio of the energy consumption of the current servers in the computer room to the total energy consumption of the computer room. These factors are all related to the current PUE of the computer room. In specific implementation, the priority sequence of the computer room is determined by these factors, that is, the server layout area is first sorted.

[0073] Specifically, the step of determining the priority sequence of the computer room according to the computing power factor, the low load ratio and the energy consumption utilization rate according to a preset rule includes:

[0074] A preset coordinate system is established with low load proportion and energy consumption utilization rate as the horizontal and vertical coordinates of the coordinate system respectively;

[0075] Mapping the computer room to the preset coordinate system according to the low load proportion and energy utilization rate of the computer room;

[0076] Taking the coordinate origin as a base point, a plurality of sector-shaped areas with a preset radius are divided to determine a plurality of priority determination areas, and the basic priority of the computer room is determined according to the priority determination area where the computer room is located.

[0077] Among them, Figure 2As shown, a preset coordinate system is established with low load proportion and energy utilization rate as the horizontal and vertical coordinates of the coordinate system respectively, and the computer room is mapped to the preset coordinate system according to the low load proportion and energy utilization rate of the computer room, which are represented by black dots in the figure, such as A1, A2, A3, A4, A5 and A6. A plurality of priority determination areas S are determined by dividing a plurality of sector areas with a preset radius with the coordinate origin as the base point, and the basic priority of the computer room is determined according to the priority determination area in which the computer room is located, wherein a plurality of priority determination areas S are set, and those in the priority determination area S are attributed to the priority, for example, in an embodiment of the present invention, a first priority determination area, a second priority determination area and a third priority determination area are set, and the outermost one is set as the first priority determination area, wherein A1 is in the first priority determination area, and the basic priority of the computer room is the first priority, wherein the preset radius can be set according to the actual situation and is not limited here.

[0078] In addition, in some optional embodiments of the present invention, the step of taking the coordinate origin as a base point and dividing a plurality of sector-shaped areas with a preset radius to determine a plurality of priority determination areas, and determining the basic priority of the computer room according to the priority determination area where the computer room is located further includes:

[0079] Obtain the computing power factor, low load ratio and energy utilization rate of the computer room within the basic priority level respectively, and determine the ranking factor of the computer room according to the computing power factor and low load ratio;

[0080] The sub-priority of the computer room is determined according to the energy consumption utilization rate and the ranking factor, and the priority sequence of the computer room is determined according to the basic priority and the sub-priority.

[0081] Among them, the sub-priority of the computer room is determined by the ranking factor of the computer room, and the priority sequence of the computer room is determined according to the basic priority and sub-priority of the computer room. The computer rooms in the first level are calculated respectively, and the priority of the computer rooms in the first level is sorted again. When the priority calculation of all computer rooms is completed, the final priority sequence is obtained. For example, the basic priority of computer room A1 is the first level, and it is in the second level in the first level. The priority of the computer room is the second level of the first level. Specifically, the calculation formula is:

[0082]

[0083] Among them, μ j is the energy utilization rate of the jth computer room, ε i is the adjustment coefficient of the i-th ranking factor, which is used to balance the values ​​of various ranking factors. ij is the value of the i-th sorting factor of the j-th computer room, that is, the computing power factor and energy utilization rate. N is a positive integer, 0≤N≤2.

[0084] It can be understood that prioritizing the computer rooms has at least the following beneficial effects:

[0085] 1. Determine the final priority ranking through multi-level priority ranking of computer rooms and server layout areas, making the allocation of tasks more reasonable;

[0086] 2. Use factors related to PUE values ​​as priority ranking indicators to make the ranking of computer rooms more accurate;

[0087] 3. By sorting the dispatch of migration tasks, since the energy consumption of other energy-consuming equipment is operated at rated power, the possibility of adjusting a certain server layout area is guaranteed while migrating tasks, further reducing the PUE value.

[0088] In summary, the method for reducing the PUE value of the intelligent computing center in the above embodiment of the present invention obtains a target server in a low-load operating state from the server, and determines the server layout area to which the target server belongs; sorts the computing task migration priority of the server layout area according to the number of target servers in the server layout area to determine the computing task migration priority of the server layout area; obtains several computing tasks outside the intelligent computing center, and migrates the computing tasks to the target servers in the current server layout area according to the computing task migration priority, that is, according to the established rules, the tasks of the servers in the server layout area are migrated in sequence, and the tasks of the servers in the server layout area with the front ranking are migrated in an orderly manner, thereby realizing the reasonable allocation of computing task migration, and solving the problem of unreasonable computing task allocation when reducing the PUE value efficiency in the prior art. In addition, since the load of the servers in each server layout area is increased first and then the load of the servers in other server layout areas is increased, the possibility of adjusting the energy-consuming equipment in the server layout area with a large number of target servers is retained, and the energy consumption is reduced by adjusting the power of the energy-consuming equipment, thereby further reducing the PUE value.

[0089] Embodiment 3

[0090] See also Figure 3 , shown is a system for reducing the PUE value of an intelligent computing center proposed in a third embodiment of the present invention, which is used to reduce the PUE value of an intelligent computing center. The intelligent computing center includes a plurality of computer rooms for arranging servers, and a plurality of server arrangement areas arranged side by side are arranged in the computer rooms. The servers are distributed in the server arrangement areas. The system includes:

[0091] An acquisition module 100 is used to acquire a target server in a low-load operation state from the servers, and determine a server deployment area to which the target server belongs;

[0092] A sorting module 200 is used to sort the computing task migration priorities of the server deployment area according to the number of the target servers in the server deployment area to determine the computing task migration priority of the server deployment area;

[0093] The migration module 300 is used to obtain several computing tasks outside the intelligent computing center, and migrate the computing tasks to the target servers in the current server deployment area according to the computing task migration priority until all the target servers in the current server deployment area are in full load operation state.

[0094] Furthermore, the system for reducing the PUE value of the intelligent computing center, wherein the system further includes:

[0095] A computing power determination module, used to obtain the remaining computing power of each server in each of the computer rooms, so as to determine the computing power factor of the computer room according to the remaining computing power of each server;

[0096] A proportion determination module, used to determine the low load proportion of the computer room according to the target server and the total number of servers in the computer room that are under low load, and determine the energy consumption utilization rate of the computer room according to the percentage proportion of the server energy consumption of the computer room to the energy consumption of the computer room;

[0097] A priority sequence determination module, used to determine the priority sequence of the computer room according to preset rules based on the computing power factor, low load ratio and energy consumption utilization rate;

[0098] The migration module is specifically used for:

[0099] When task migration is required, the target computer room that currently needs to perform task migration is determined according to the priority sequence of the computer room, and the current server layout area is determined from the target computer room according to the computing task migration priority, and the computing task is migrated to the target server in the current server layout area.

[0100] Further, in some optional embodiments of the present invention, the priority sequence determination module includes:

[0101] Establishing a unit for establishing a preset coordinate system with low load proportion and energy consumption utilization rate as the horizontal coordinate and vertical coordinate of the coordinate system respectively;

[0102] A mapping unit, configured to map the computer room to the preset coordinate system according to the low load ratio and energy utilization rate of the computer room;

[0103] The division unit is used to divide the coordinate origin as a base point and determine multiple priority determination areas by dividing multiple sector-shaped areas with a preset radius, and determine the basic priority of the computer room according to the priority determination area where the computer room is located.

[0104] Furthermore, in some optional embodiments of the present invention, the priority sequence determination module further includes:

[0105] A ranking factor determination unit, used to respectively obtain the computing power factor, low load ratio and energy consumption utilization rate of the computer room within the basic priority level, and determine the ranking factor of the computer room according to the computing power factor and low load ratio;

[0106] The priority sequence determination unit is used to determine the sub-priority of the computer room according to the energy consumption utilization rate and the sorting factor, and determine the priority sequence of the computer room according to the basic priority and the sub-priority.

[0107] Further, in some optional embodiments of the present invention, the calculation formula for determining the sub-priority of the computer room according to the energy consumption utilization rate and the ranking factor is:

[0108]

[0109] Among them, μ j is the energy utilization rate of the jth computer room, ε i is the adjustment coefficient of the ith ranking factor, L ij is the value of the i-th sorting factor of the j-th computer room, N is a positive integer, 0≤N≤2.

[0110] Furthermore, in some optional embodiments of the present invention, the system further includes:

[0111] The adjustment module is used to determine the target server deployment area where the number of target servers is lower than the preset number after the computing task allocation is completed, and adjust the refrigeration system of the target server deployment area to reduce the energy consumption of the refrigeration system.

[0112] Furthermore, in some optional embodiments of the present invention, the system further includes:

[0113] A judgment module, used to judge whether the target server layout area meets the cooling system adjustment conditions;

[0114] If so, the step of adjusting the refrigeration system of the target server deployment area to reduce the energy consumption of the refrigeration system is performed.

[0115] The functions or operation steps implemented when the above modules are executed are substantially the same as those in the above method embodiments, and will not be repeated here.

[0116] Embodiment 4

[0117] Another aspect of the present invention further provides a readable storage medium having a computer program stored thereon, and when the program is executed by a processor, the steps of the method described in any one of the above embodiments 1 to 2 are implemented.

[0118] Embodiment 5

[0119] Another aspect of the present invention provides an electronic device, which includes a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor implements the steps of the method described in any one of the above-mentioned embodiments one to two when executing the program.

[0120] The technical features of the above embodiments can be combined arbitrarily. To make the description concise, not all possible combinations of the technical features in the above embodiments are described. However, as long as there is no contradiction in the combination of these technical features, they should be considered to be within the scope of this specification.

[0121] Those skilled in the art will appreciate that the logic and / or steps represented in the flowchart or otherwise described herein, for example, may be considered as an ordered list of executable instructions for implementing logical functions, and may be embodied in any computer-readable medium for use by an instruction execution system, device or apparatus (such as a computer-based system, a system including a processor, or other system that can fetch instructions from an instruction execution system, device or apparatus and execute instructions), or in conjunction with such instruction execution systems, devices or apparatuses. For purposes of this specification, a "computer-readable medium" may be any device that can contain, store, communicate, propagate or transmit a program for use by an instruction execution system, device or apparatus, or in conjunction with such instruction execution systems, devices or apparatuses.

[0122] More specific examples of computer-readable media (a non-exhaustive list) include the following: an electrical connection with one or more wires (electronic device), a portable computer disk case (magnetic device), a random access memory (RAM), a read-only memory (ROM), an erasable and programmable read-only memory (EPROM or flash memory), an optical fiber device, and a portable compact disk read-only memory (CDROM). In addition, the computer-readable medium may even be a paper or other suitable medium on which the program is printed, since the program may be obtained electronically, for example, by optically scanning the paper or other medium, followed by editing, deciphering or, if necessary, processing in another suitable manner, and then stored in a computer memory.

[0123] It should be understood that the various parts of the present invention can be implemented by hardware, software, firmware or a combination thereof. In the above-mentioned embodiments, a plurality of steps or methods can be implemented by software or firmware stored in a memory and executed by a suitable instruction execution system. For example, if implemented by hardware, as in another embodiment, it can be implemented by any one of the following technologies known in the art or a combination thereof: a discrete logic circuit having a logic gate circuit for implementing a logic function for a data signal, a dedicated integrated circuit having a suitable combination of logic gate circuits, a programmable gate array (PGA), a field programmable gate array (FPGA), etc.

[0124] In the description of this specification, the description with reference to the terms "one embodiment", "some embodiments", "examples", "specific examples", or "some examples" means that the specific features, structures, materials or characteristics described in conjunction with the embodiment or example are included in at least one embodiment or example of the present invention. In this specification, the schematic representation of the above terms does not necessarily refer to the same embodiment or example. Moreover, the specific features, structures, materials or characteristics described may be combined in any one or more embodiments or examples in a suitable manner.

[0125] The above-mentioned embodiments only express several implementation methods of the present invention, and the description thereof is relatively specific and detailed, but it cannot be understood as limiting the scope of the patent of the present invention. It should be pointed out that, for ordinary technicians in this field, several variations and improvements can be made without departing from the concept of the present invention, which all belong to the protection scope of the present invention. Therefore, the protection scope of the patent of the present invention shall be subject to the attached claims.

Claims

1. A method for reducing the PUE value of the intelligent computing center. It is characterized in that The method is used to reduce the PUE value of an intelligent computing center, wherein the intelligent computing center includes a plurality of computer rooms for arranging servers, wherein a plurality of server arrangement areas arranged side by side are arranged in the computer rooms, and wherein the servers are distributed in the server arrangement areas, and wherein the method includes: Acquire a target server in a low-load operation state from the servers, and determine a server deployment area to which the target server belongs; According to the number of the target servers in the server deployment area, the server deployment area is prioritized for computing task migration to determine the computing task migration priority of the server deployment area; Acquire a number of computing tasks outside the intelligent computing center, and migrate the computing tasks to the target servers in the current server deployment area according to the computing task migration priority, until all the target servers in the current server deployment area are in full load operation state; The step of obtaining a plurality of computing tasks outside the intelligent computing center and migrating the computing tasks to the target servers in the current server deployment area according to the computing task migration priority until all the target servers in the current server deployment area are in a full load operation state also includes: Obtaining the remaining computing power of each server in each of the computer rooms, so as to determine the computing power factor of the computer room according to the remaining computing power of each server; Determine the low load ratio of the computer room according to the target server and the total number of servers in the computer room that are under low load, and determine the energy utilization rate of the computer room according to the percentage ratio of the server energy consumption of the computer room to the energy consumption of the computer room; Determine the priority sequence of the computer room according to the computing power factor, the low load ratio and the energy consumption utilization rate according to preset rules; The step of obtaining a plurality of computing tasks outside the intelligent computing center and migrating the computing tasks to the target servers in the current server deployment area according to the computing task migration priority until all the target servers in the current server deployment area are in a full load operation state comprises: When task migration is required, the target computer room that currently needs to perform task migration is determined according to the priority sequence of the computer room, and the current server layout area is determined from the target computer room according to the computing task migration priority, and the computing task is migrated to the target server in the current server layout area; The step of determining the priority sequence of the computer room according to the computing power factor, the low load ratio and the energy consumption utilization rate according to a preset rule comprises: A preset coordinate system is established with low load proportion and energy consumption utilization rate as the horizontal and vertical coordinates of the coordinate system respectively; Mapping the computer room to the preset coordinate system according to the low load proportion and energy utilization rate of the computer room; Taking the coordinate origin as the base point, a plurality of sector-shaped areas with a preset radius are divided to determine a plurality of priority determination areas, and the basic priority of the computer room is determined according to the priority determination area in which the computer room is located.

2. The method for reducing the PUE value of the intelligent computing center according to claim 1, It is characterized in that The step of dividing a plurality of sector-shaped areas with a preset radius based on the coordinate origin as a base point to determine a plurality of priority determination areas, and determining the basic priority of the computer room according to the priority determination area where the computer room is located further includes: Obtain the computing power factor, low load ratio and energy utilization rate of the computer room within the basic priority level respectively, and determine the ranking factor of the computer room according to the computing power factor and low load ratio; The sub-priority of the computer room is determined according to the energy consumption utilization rate and the ranking factor, and the priority sequence of the computer room is determined according to the basic priority and the sub-priority.

3. The method for reducing the PUE value of the intelligent computing center according to claim 2, It is characterized in that The calculation formula for determining the secondary priority of the computer room according to the energy consumption utilization rate and the ranking factor is: ; in, is the energy utilization rate of the jth computer room, is the adjustment coefficient of the i-th ranking factor, is the value of the i-th sorting factor of the j-th computer room, N is a positive integer, 0≤N≤2.

4. The method for reducing the PUE value of the intelligent computing center according to claim 1, It is characterized in that Each of the server layout areas is correspondingly provided with a separate cooling system, and the method further comprises: After the computing task allocation is completed, a target server deployment area in which the number of target servers is greater than a preset number is determined, and a refrigeration system of the target server deployment area is adjusted to reduce energy consumption of the refrigeration system.

5. The method for reducing the PUE value of the intelligent computing center according to claim 4, It is characterized in that Before the step of adjusting the refrigeration system of the target server arrangement area to reduce the energy consumption of the refrigeration system, the step further includes: Determining whether the target server layout area meets the refrigeration system adjustment conditions; If so, the step of adjusting the refrigeration system of the target server deployment area to reduce the energy consumption of the refrigeration system is performed.

6. A system for reducing the PUE value of the intelligent computing center, It is characterized in that Used to reduce the PUE value of an intelligent computing center, the intelligent computing center includes a plurality of computer rooms for arranging servers, the computer rooms are provided with a plurality of server arrangement areas arranged side by side, the servers are distributed in the server arrangement areas, and the method for reducing the PUE value of an intelligent computing center according to any one of claims 1 to 5 is implemented, the system comprising: An acquisition module, used for acquiring a target server in a low-load operation state from the servers, and determining a server deployment area to which the target server belongs; A sorting module, configured to sort the computing task migration priorities of the server deployment area according to the number of the target servers in the server deployment area, so as to determine the computing task migration priorities of the server deployment area; A migration module is used to obtain several computing tasks outside the intelligent computing center, and migrate the computing tasks to the target servers in the current server deployment area according to the computing task migration priority until all the target servers in the current server deployment area are in full load operation state.

7. A readable storage medium having a computer program stored thereon, It is characterized in that When the program is executed by a processor, the steps of the method according to any one of claims 1 to 6 are implemented.

8. An electronic device, It is characterized in that The method comprises a memory, a processor and a computer program stored in the memory and executable on the processor, wherein when the processor executes the program, the steps of the method according to any one of claims 1 to 6 are implemented.

Citation Information

Cited By

  • Distributed intelligent calculation center energy consumption control method and system for wind power generation coupling in mountainous area

    CN121727134A