CN117666741A - Server heat dissipation control method and system - Google Patents
Server heat dissipation control method and system Download PDFInfo
- Publication number
- CN117666741A CN117666741A CN202410150492.6A CN202410150492A CN117666741A CN 117666741 A CN117666741 A CN 117666741A CN 202410150492 A CN202410150492 A CN 202410150492A CN 117666741 A CN117666741 A CN 117666741A
- Authority
- CN
- China
- Prior art keywords
- heat dissipation
- regulation
- scheme
- control
- server
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Granted
Links
- 230000017525 heat dissipation Effects 0.000 title claims abstract description 322
- 238000000034 method Methods 0.000 title claims abstract description 46
- 230000033228 biological regulation Effects 0.000 claims abstract description 192
- 230000008859 change Effects 0.000 claims abstract description 68
- 230000003044 adaptive effect Effects 0.000 claims abstract description 37
- 238000012549 training Methods 0.000 claims abstract description 19
- 230000008844 regulatory mechanism Effects 0.000 claims abstract description 13
- 230000003213 activating effect Effects 0.000 claims abstract description 9
- 230000002860 competitive effect Effects 0.000 claims abstract description 7
- 238000005457 optimization Methods 0.000 claims description 51
- 230000005855 radiation Effects 0.000 claims description 18
- 230000008569 process Effects 0.000 claims description 12
- 238000004891 communication Methods 0.000 claims description 9
- 238000013507 mapping Methods 0.000 claims description 8
- 238000012545 processing Methods 0.000 claims description 8
- 230000011218 segmentation Effects 0.000 claims description 8
- 230000004913 activation Effects 0.000 claims description 5
- 230000003247 decreasing effect Effects 0.000 claims description 4
- 238000012216 screening Methods 0.000 claims description 4
- 230000002457 bidirectional effect Effects 0.000 claims description 3
- 230000003993 interaction Effects 0.000 claims description 3
- 230000007246 mechanism Effects 0.000 claims description 2
- 230000009123 feedback regulation Effects 0.000 abstract 2
- 230000017986 regulation of heat dissipation Effects 0.000 abstract 1
- 230000001105 regulatory effect Effects 0.000 description 6
- 238000012360 testing method Methods 0.000 description 5
- 238000001816 cooling Methods 0.000 description 4
- 238000011161 development Methods 0.000 description 4
- 230000018109 developmental process Effects 0.000 description 4
- 230000000694 effects Effects 0.000 description 4
- 238000010276 construction Methods 0.000 description 3
- 230000001276 controlling effect Effects 0.000 description 3
- 230000000630 rising effect Effects 0.000 description 2
- 230000005540 biological transmission Effects 0.000 description 1
- 238000010586 diagram Methods 0.000 description 1
- 238000005516 engineering process Methods 0.000 description 1
- 239000007788 liquid Substances 0.000 description 1
- 238000010801 machine learning Methods 0.000 description 1
- 239000000463 material Substances 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 230000035772 mutation Effects 0.000 description 1
- 238000003062 neural network model Methods 0.000 description 1
- 230000009467 reduction Effects 0.000 description 1
- 238000004904 shortening Methods 0.000 description 1
- 230000001360 synchronised effect Effects 0.000 description 1
- 238000012546 transfer Methods 0.000 description 1
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F1/00—Details not covered by groups G06F3/00 - G06F13/00 and G06F21/00
- G06F1/16—Constructional details or arrangements
- G06F1/20—Cooling means
-
- H—ELECTRICITY
- H05—ELECTRIC TECHNIQUES NOT OTHERWISE PROVIDED FOR
- H05K—PRINTED CIRCUITS; CASINGS OR CONSTRUCTIONAL DETAILS OF ELECTRIC APPARATUS; MANUFACTURE OF ASSEMBLAGES OF ELECTRICAL COMPONENTS
- H05K7/00—Constructional details common to different types of electric apparatus
- H05K7/20—Modifications to facilitate cooling, ventilating, or heating
- H05K7/20709—Modifications to facilitate cooling, ventilating, or heating for server racks or cabinets; for data centers, e.g. 19-inch computer racks
- H05K7/20836—Thermal management, e.g. server temperature control
-
- Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
- Y02—TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
- Y02D—CLIMATE CHANGE MITIGATION TECHNOLOGIES IN INFORMATION AND COMMUNICATION TECHNOLOGIES [ICT], I.E. INFORMATION AND COMMUNICATION TECHNOLOGIES AIMING AT THE REDUCTION OF THEIR OWN ENERGY USE
- Y02D10/00—Energy efficient computing, e.g. low power processors, power management or thermal management
Landscapes
- Engineering & Computer Science (AREA)
- General Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- Computer Hardware Design (AREA)
- Thermal Sciences (AREA)
- Microelectronics & Electronic Packaging (AREA)
- Human Computer Interaction (AREA)
- General Physics & Mathematics (AREA)
- Feedback Control In General (AREA)
Abstract
The invention provides a server heat dissipation control method and a system, which relate to the technical field of intelligent control, wherein the method comprises the following steps: setting a multi-stage temperature change grade determining mode regulation mechanism, supervising and training a self-adaptive heat dissipation regulation model, reading heat dissipation record data and determining a target temperature change grade, activating a target regulation branch mapped to the target temperature change grade, if the target regulation branch is an advanced regulation branch, analyzing and determining the heat dissipation regulation scheme to perform scheme expansion and competitive assimilation optimizing to determine a target heat dissipation scheme, feeding the target heat dissipation scheme back to a server heat dissipation system, returning heat dissipation feedback data, generating a feedback regulation command if the target heat dissipation scheme does not meet an expected heat dissipation standard, and transmitting the feedback regulation command and the heat dissipation feedback data back to the self-adaptive heat dissipation regulation model, so that heat dissipation control analysis is repeatedly performed, the technical problems of insufficient heat dissipation control of a server and low operation efficiency of the server in the prior art are solved, the adaptive regulation of heat dissipation of the server is realized, and the operation efficiency of the server is further improved.
Description
Technical Field
The invention relates to the technical field of intelligent control, in particular to a server heat dissipation control method and system.
Background
With the development of the cloud computing industry, particularly the development of servers, more and more data centers are built under the background of the development of the cloud computing industry, and the servers are used as the most core equipment of the data centers, so that the high performance, the high availability and the high cost performance of the servers become important indexes for measuring the quality of the servers. Because of the limited volume of the server, whether a plurality of high-power electronic components run for a long time and under high load can timely transfer the heat generated by the electronic components to the outside is directly related to the running stability of the server. Therefore, the heat dissipation problem of the server becomes a big obstacle restricting the development of the server, and the heat dissipation management and control of the server are insufficient in the prior art, so that the server has the technical problem of low operation efficiency.
Disclosure of Invention
The application provides a server heat dissipation control method and a server heat dissipation control system, which are used for solving the technical problem that the operation efficiency of a server is low due to insufficient management and control of the heat dissipation of the server in the prior art.
In view of the above problems, the present application provides a server heat dissipation control method and system.
In a first aspect, the present application provides a server heat dissipation control method, where the method includes: setting multi-stage temperature change grades, carrying out adaptive configuration of a heat dissipation regulation mode, and determining a mode regulation mechanism, wherein the heat dissipation regulation mode comprises a conventional regulation mode and an advanced regulation mode, and each stage of temperature change grade comprises a temperature rise rate threshold value and a critical temperature value; the mode regulation mechanism is combined, an adaptive heat dissipation regulation model is supervised and trained, and the adaptive heat dissipation regulation model is in communication connection with a server heat dissipation system and comprises a parallel conventional regulation branch and an advanced regulation branch; reading heat dissipation record data of a radiator, determining a target temperature change grade, and activating a target regulation branch mapped to the target temperature change grade in the self-adaptive heat dissipation regulation model; if the target regulation branch is the advanced regulation branch, analyzing and determining a heat dissipation regulation scheme, and performing scheme expansion and competitive assimilation optimizing to determine a target heat dissipation scheme, wherein a heat dissipation system with distributed weights is identified, software optimization and CPU frequency modulation are used as heat dissipation regulation analysis standards; feeding back the target heat dissipation scheme to the server heat dissipation system, performing server heat dissipation management and control based on a radiator, and returning heat dissipation feedback data; judging whether the heat radiation feedback data meets an expected heat radiation standard, if not, generating a feedback adjustment instruction, wherein the expected heat radiation standard has a tolerant interval based on technical constraint and scene influence; and the feedback adjustment instruction and the heat dissipation feedback data are returned to the self-adaptive heat dissipation regulation model, and heat dissipation control analysis is repeatedly performed.
In a second aspect, the present application provides a server heat dissipation control system, the system comprising: one or more technical solutions provided in the present application have at least the following technical effects or advantages: the configuration module is used for setting multi-stage temperature change grades, carrying out adaptive configuration of a heat dissipation regulation mode and determining a mode regulation mechanism, wherein the heat dissipation regulation mode comprises a conventional regulation mode and an advanced regulation mode, and each stage of temperature change grade comprises a temperature rise rate threshold value and a critical temperature value; the training module is used for combining the mode regulation and control mechanism, supervising and training an adaptive heat dissipation regulation and control model, and the adaptive heat dissipation regulation and control model is in communication connection with a server heat dissipation system and comprises a parallel conventional regulation and control branch and an advanced regulation and control branch; the activation module is used for reading heat dissipation record data of the radiator and determining a target temperature change grade, and activating a target regulation branch mapped to the target temperature change grade in the self-adaptive heat dissipation regulation model; the optimizing module is used for analyzing and determining a heat dissipation regulation and control scheme if the target regulation and control branch is the advanced regulation and control branch, and performing scheme expansion and competitive assimilation optimizing to determine a target heat dissipation scheme, wherein a heat dissipation system with distributed weights is identified, software optimization and CPU frequency modulation are used as heat dissipation regulation and control analysis standards; the management and control module is used for feeding back the target heat dissipation scheme to the server heat dissipation system, carrying out server heat dissipation management and control based on a radiator and returning heat dissipation feedback data; the first judging module is used for judging whether the heat radiation feedback data meet expected heat radiation standards, if not, a feedback adjustment instruction is generated, and the expected heat radiation standards have a tolerance interval based on technical constraint and scene influence; and the control analysis module is used for transmitting the feedback adjustment instruction and the heat dissipation feedback data back to the self-adaptive heat dissipation regulation and control model, and repeatedly carrying out heat dissipation control analysis.
The application provides a server heat dissipation control method and a server heat dissipation control system, relates to the technical field of intelligent control, solves the technical problem that the server heat dissipation is not enough to control in the prior art, and the running efficiency of the server is low, and realizes the adaptive control of the heat dissipation of the server, so that the running efficiency of the server is improved.
Drawings
Fig. 1 is a schematic flow chart of a server heat dissipation control method.
Fig. 2 is a schematic structural diagram of a server heat dissipation control system provided in the present application.
Reference numerals illustrate: the system comprises a configuration module 1, a training module 2, an activating module 3, an optimizing module 4, a management and control module 5, a first judging module 6 and a control and analysis module 7.
Detailed Description
The method and the system for controlling the heat dissipation of the server are used for solving the technical problems of low operation efficiency of the server caused by insufficient management and control of the heat dissipation of the server in the prior art, and realizing the adaptive regulation and control of the heat dissipation of the server.
Example 1
As shown in fig. 1, an embodiment of the present application provides a server heat dissipation control method, which includes:
step A100: setting multi-stage temperature change grades, carrying out adaptive configuration of a heat dissipation regulation mode, and determining a mode regulation mechanism, wherein the heat dissipation regulation mode comprises a conventional regulation mode and an advanced regulation mode, and each stage of temperature change grade comprises a temperature rise rate threshold value and a critical temperature value;
in this application, the heat dissipation control method of the server provided in this application is applied to a heat dissipation control system of the server, so that in order to ensure the accuracy of heat dissipation control of the server, it is required to set multiple temperature change levels according to the temperature change amplitude in the running process of the server, where the multiple temperature change levels may include level a, level b and level c, level a may be 22±2 ℃, level b may be 15 to 30 ℃ and level c may be 10 to 35 ℃, on this basis, the adaptive configuration of the heat dissipation control mode is performed on the server, so as to determine a mode control extremum, and in the heat dissipation control mode, a conventional control mode and a high level control mode may be included, where the conventional control mode refers to performing conventional heat dissipation processing if the temperature of the server is lower in a controllable range, that is, in a normal state, where the high level control mode refers to performing more cumbersome cooling processing on the server when the temperature of the server is higher, and may further include a rate threshold and a critical temperature value in the temperature change level of each level, where the rate threshold is the critical temperature value is higher than the critical temperature value in the running range, and the critical temperature value is higher than the critical temperature value of the server, and the critical temperature value is controlled as a reference value when the server is in the critical temperature raising process.
Step A200: the mode regulation mechanism is combined, an adaptive heat dissipation regulation model is supervised and trained, and the adaptive heat dissipation regulation model is in communication connection with a server heat dissipation system and comprises a parallel conventional regulation branch and an advanced regulation branch;
in the application, by combining the supervision training self-adaptive heat dissipation regulation model with the determined mode regulation mechanism, it is meant that the temperature change level and the heat dissipation regulation mode in the mode regulation mechanism are combined as training data, when the temperature is lower, namely the temperature rising rate and the temperature value are lower, the conventional regulation mode is used for dissipating heat from the server, when the temperature is higher, namely the temperature rising rate and the temperature value are synchronous and higher, the heat dissipation efficiency is accelerated in order to avoid the mode of influencing the actual operation based on CPU (Central processing Unit) frequency reduction and the like, scheme optimizing is carried out, the self-adaptive heat dissipation regulation model is in communication connection with the server heat dissipation system, and comprises parallel conventional regulation branches and advanced regulation branches, the self-adaptive heat dissipation regulation model is a neural network model which can be subjected to self-iterative optimization continuously in machine learning, the parallel conventional regulation branches and advanced regulation branches are all independent operation branches, the self-adaptive activation is carried out along with the continuous change of the heat dissipation regulation demand, further, the self-adaptive heat dissipation regulation model is subjected to bidirectional interaction of data between the self-adaptive heat dissipation regulation model and the server heat dissipation system, the self-adaptive heat dissipation regulation model is obtained through a training data set and a supervision data set, and the supervision data set is a data set corresponding to the training data set one by one.
Further, the construction process of the self-adaptive heat dissipation regulation model is as follows: inputting each group of training data in the training data set into the self-adaptive heat dissipation regulation model, performing output supervision adjustment of the self-adaptive heat dissipation regulation model through supervision data corresponding to the group of training data, finishing the current group of training when the output result of the self-adaptive heat dissipation regulation model is consistent with the supervision data, finishing all training data in the training data set, and finishing the training of the self-adaptive heat dissipation regulation model.
In order to ensure the convergence and accuracy of the adaptive heat dissipation regulation model, the convergence process may be that when the output data in the adaptive heat dissipation regulation model is converged to one point, the convergence is performed when a certain value is close, and the accuracy may be that the test processing of the adaptive heat dissipation regulation model is performed through the test data set, for example, the test accuracy may be set to 80%, and when the test accuracy of the test data set meets 80%, the construction of the adaptive heat dissipation regulation model is completed, so as to ensure the heat dissipation control of the server.
Step A300: reading heat dissipation record data of a radiator, determining a target temperature change grade, and activating a target regulation branch mapped to the target temperature change grade in the self-adaptive heat dissipation regulation model;
further, step a300 of the present application further includes:
step a310: determining a segmentation baseline based on a heat dissipation regulation mode, wherein the segmentation baseline comprises a reference temperature rise rate and a reference temperature value;
step A320: classifying and attributing the multi-stage temperature change grades based on the segmentation baseline, determining a first temperature change grade and a second temperature change grade, wherein the multi-stage temperature change grades comprise at least three stages and present an increasing trend;
step a330: and performing mapping configuration of the first temperature change level and the conventional regulation mode and mapping configuration of the second temperature change level and the advanced regulation mode.
In the application, in order to improve the control efficiency in the process of radiating the server, the radiating record data contained in the radiator needs to be read, which means the radiating data in the time zone bordering the current time, meanwhile, the radiating record can contain the radiating rate, radiating duration and the like to determine the temperature change of the server, and the radiating rate, radiating duration and the like, and the temperature change is recorded as a target temperature change level, the target regulation branch of the mapping and the target temperature change level is activated in the adaptive radiating regulation model, the mapped target regulation branch of the target temperature change level is attributed to the multistage temperature change level by dividing the multistage temperature change level on the basis of a division baseline of the radiating regulation mode, the reference temperature rise rate means the temperature rise rate and the reference temperature data value of the server in the normal running state, so that the multistage temperature change level can be divided into a first temperature change level and a second temperature change level on the basis of the division baseline, the first temperature change level and the second temperature change level respectively comprise a trend, the first stage, the second stage and the third stage are compared, the first stage temperature change level is gradually increased on the basis of the normal running state, and the second stage temperature change is configured to realize the high-stage regulation and the normal temperature regulation mode.
Step A400: if the target regulation branch is the advanced regulation branch, analyzing and determining a heat dissipation regulation scheme, and performing scheme expansion and competitive assimilation optimizing to determine a target heat dissipation scheme, wherein a heat dissipation system with distributed weights is identified, software optimization and CPU frequency modulation are used as heat dissipation regulation analysis standards;
further, step a400 of the present application further includes:
step A410: wherein the advanced regulation branch is connected with an industrial database channel;
step a420: searching and determining the initialized heat dissipation regulation scheme in the industrial database by taking the heat dissipation record data as an index;
step a430: determining the degree of freedom of regulation and control based on the heat dissipation system, software optimization and CPU frequency modulation, and constructing diversified optimizing space;
step a440: and taking the control freedom degree as constraint, executing expansion optimization based on the heat dissipation control scheme in the optimizing space, and determining the target heat dissipation scheme.
Further, step a440 of the present application includes:
step a441: taking the regulation and control degrees of freedom as constraint, carrying out random disturbance of the heat dissipation regulation and control scheme in the optimizing space, and determining an expansion scheme set;
step a442: the method comprises the steps of carrying out positive sequence ordering of an expansion scheme set based on scheme superiority and inferiority, determining a preset dividing ratio, dividing the expansion scheme set to determine a plurality of groups of expansion schemes, wherein the preset dividing ratio is sequentially decreased, and the plurality of groups of expansion schemes comprise at least three groups;
step A443: optimizing and determining the target heat dissipation scheme based on the plurality of groups of expansion schemes;
step a444: wherein performing inter-group optimization based on the plurality of groups of expansion schemes comprises:
step a445: determining an optimal solution of a group of expansion schemes, taking the optimal solution as an optimization direction, optimizing two groups of expansion schemes, and determining a first optimization scheme set;
step a446: taking the group of expansion schemes and the first optimization scheme set as optimization directions, carrying out random matching and optimization on the three groups of expansion schemes, and determining a second optimization scheme set;
step a447: and sequentially performing iterative optimization, checking the group of expansion schemes, the first optimization scheme set and the second optimization scheme set, and selecting an optimal scheme as the target heat dissipation scheme.
In the application, in order to more accurately perform heat dissipation control on a server, a heat dissipation scheme is required to be specified first, when a target regulation and control branch mapping in an adaptive heat dissipation regulation and control model is configured as an advanced regulation and control branch, the temperature of the server is higher than the temperature under the conventional condition, and the temperature value is gradually increased, so that analysis and determination on the heat dissipation regulation and control scheme are realized, the advanced regulation and control branch is connected with an industrial database channel, the industrial database channel is a data solution space for storing various data records of the server in the operation process, further, heat dissipation record data is used as index data, traversal access retrieval of industrial data one by one is performed in an industrial database, the industrial data matched with the heat dissipation record data is extracted, and on the basis, the initialized heat dissipation regulation and control scheme is determined, then the heat dissipation system, the software optimization and the regulating degree of freedom of CPU frequency modulation are used as reference basic data, the heat dissipation system absorbs heat and rapidly transmits the heat to the surface of the heat sink through contacting with a heat source in a server, the software optimization means that the server can handle more service requests by increasing the self resource scale, the regulating degree of freedom of the CPU frequency modulation means that when the load of the CPU is higher, the frequency of the CPU can be automatically increased to improve the performance of the CPU, thereby constructing a diversified optimizing space, further, the regulating degree of freedom is used as an optimizing constraint boundary, the expansion optimizing based on a heat dissipation regulating scheme is executed in the constructed diversified optimizing space, then the regulating degree of freedom is used as a constraint, the random disturbance of the heat dissipation regulating scheme is carried out in the diversified optimizing space, the random disturbance means that the time is hidden when the optimizing space exists, the negligence is small, the change rule of the negligence cannot be described by a random interference item with a certain determined functional relation, and the determined random interference item is extracted and summarized and then is recorded as an expansion scheme set.
Further, the positive sequence ordering is carried out on the expansion scheme set by judging the superiority of the schemes, namely, the superiority of the schemes is determined according to the positive influence degree of random disturbance on heat dissipation control, when the positive influence degree is higher, the schemes are better, otherwise, the positive influence degree is lower, the schemes are worse, the positive sequence ordering operation is carried out on the schemes in the expansion scheme set according to the priority, on the basis, the preset division proportion is determined, the division expansion scheme set is determined into a plurality of groups of expansion schemes, wherein the preset division proportion is sequentially decreased, the decreasing proportion can be 5%, 10%, 85% and the like, the scheme with the inferior rear part is randomly determined into an optimization direction for adjustment, so that the optimal randomness of the inferior state schemes is improved, the mutation rate is improved, and the plurality of groups of expansion schemes have a corresponding relation with the multi-stage temperature change level, the method comprises selecting and determining optimal solutions of one group of expansion schemes randomly in multiple groups of expansion schemes as optimization directions, namely data optimization directions, optimizing two groups of expansion schemes to determine a first optimization scheme set, wherein the two groups of expansion schemes are different from the one group of expansion schemes, the optimal solutions of the two groups of expansion schemes are smaller than the one group of expansion schemes, further, the one group of expansion schemes and the first optimization scheme set are used as optimization directions, the solutions in the three groups of expansion schemes are randomly matched and the optimized solutions are recorded as a second optimization scheme set, then iterative optimization is sequentially carried out, and a group of expansion schemes, a first optimization scheme set and a second optimization scheme set are checked at the same time, so that the accuracy and the integrity of the schemes are ensured to avoid errors, information transmission errors are prevented, the optimal scheme is selected as a target heat dissipation scheme to be output, and the effect of limiting heat dissipation control on a server is realized.
Step A500: feeding back the target heat dissipation scheme to the server heat dissipation system, performing server heat dissipation management and control based on a radiator, and returning heat dissipation feedback data;
further, step a500 of the present application further includes:
step A510: if the target regulation branch is a conventional regulation branch, determining a conventional heat dissipation flow;
step A520: for the heat dissipation record data, performing optional node locking and self-adaptive adjustment on the conventional heat dissipation process in the conventional regulation branch to serve as the target heat dissipation scheme;
step a530: and feeding the target heat dissipation scheme back to the server heat dissipation system to perform server heat dissipation management and control and feedback analysis.
In this application, in order to ensure the heat dissipation efficiency in the heat dissipation process of the server according to the above-determined heat dissipation scheme, it is necessary to feed back the target heat dissipation scheme to the server heat dissipation system to perform heat dissipation management and control on the server, first, it is determined whether the target regulation branch in the adaptive heat dissipation regulation model in communication connection with the server heat dissipation system is a conventional regulation branch, when the target regulation branch is the conventional regulation branch, the conventional heat dissipation process may be determining the conventional heat dissipation process on the server, and further, for the heat dissipation record data, optional node locking and adaptive adjustment are performed on the conventional heat dissipation process in the conventional regulation branch, for example, when the server cannot reduce the server to a normal temperature only by performing heat dissipation operation on the fan, after the heat dissipation node is locked, the target heat dissipation scheme may be used as the target heat dissipation scheme by selectively opening the heat dissipation fin and/or the liquid cooling operation, and simultaneously feeding back the target scheme to the server heat dissipation system, performing heat dissipation management and feedback analysis on the server, and generating data for the server feedback data, so as to perform feedback data for the heat dissipation data for the server to perform control later control and reference on the heat dissipation data.
Step A600: judging whether the heat radiation feedback data meets an expected heat radiation standard, if not, generating a feedback adjustment instruction, wherein the expected heat radiation standard has a tolerant interval based on technical constraint and scene influence;
in the application, in order to ensure the heat dissipation effect after the heat dissipation operation is performed on the server, the heat dissipation feedback data and the expected heat dissipation standard are required to be compared and judged, the expected heat dissipation standard is obtained by jointly judging the heat dissipation trend and the heat dissipation temperature value of the server, the expected heat dissipation standard is in a tolerance interval which has influence on heat dissipation in a scene where the heat dissipation operation constraint limit and the server are located based on the heat dissipation operation technology constraint, when the heat dissipation feedback data meets the expected heat dissipation standard, the current heat dissipation operation of the server is considered to be capable of effectively cooling the server, when the heat dissipation feedback data does not meet the expected heat dissipation standard, the current heat dissipation operation of the server is considered to be incapable of effectively cooling the server, a feedback adjustment instruction is generated, the current heat dissipation operation of the server is fed back and the heat dissipation is adjusted and strengthened through the feedback adjustment instruction, and the accuracy of heat dissipation control of the server is improved.
And step A700, the feedback adjustment instruction and the heat dissipation feedback data are transmitted back to the self-adaptive heat dissipation regulation model, and heat dissipation control analysis is repeated.
Further, step a700 of the present application further includes:
step a710: setting a heat dissipation base line based on normal operation and control of a server;
step A720: calling a time zone heat dissipation record in a preset time limit, performing over-frequency regulation and control analysis by combining the heat dissipation base line, screening the over-frequency heat dissipation record and calculating the over-frequency duty ratio;
step a730: judging whether the over-frequency duty ratio meets a threshold value standard or not, and if so, generating a hardware configuration instruction;
step a740: and executing the hardware optimization processing of the radiator along with the receiving of the hardware configuration instruction.
In the application, in order to improve the heat dissipation control efficiency of the server, when a feedback adjustment instruction is generated in a heat dissipation system of the server, the feedback adjustment instruction and heat dissipation feedback data are combined and then transmitted back to an adaptive heat dissipation regulation model in communication connection with the system, the control analysis of repeated heat dissipation is performed on the server, namely, firstly, a heat dissipation base line of normal operation control of the server is set, the heat dissipation base line of normal operation control of the server is set according to heat dissipation data when the temperature of the server is at a normal operation temperature in a historical period, further, a time zone heat dissipation record of the server in a preset time limit is called, namely, the time corresponding to different heat dissipation data in the preset time limit is combined with the heat dissipation base line to perform over-frequency regulation analysis at the same time, namely, by increasing the memory frequency of the server or shortening the memory time sequence, the method comprises the steps of improving the memory reading and writing speed so as to reduce the heat added by the operation of a server, screening the super-frequency heat dissipation record, calculating the super-frequency duty ratio existing in the server, namely, determining the super-frequency duty ratio after the super-frequency heat dissipation times in the super-frequency heat dissipation record and the heat dissipation times in the heat dissipation record are made a quotient, then judging whether the super-frequency duty ratio meets a threshold standard or not, wherein the threshold standard is that the super-frequency heat dissipation time duty ratio of the server in a historical period is used for judging that the super-frequency duty ratio meets the threshold standard, generating a hardware configuration instruction, executing hardware optimization processing of the heat dissipation device on the server through the hardware configuration instruction, and exemplarily, executing operation of changing a heat dissipation material with high heat conductivity coefficient on the server, namely, adding a fan, adding fins and the like, so as to ensure better heat dissipation control on the server in the later period.
In summary, the method for controlling the heat dissipation of the server provided by the embodiment of the application at least includes the following technical effects, so that the heat dissipation of the server is adaptively controlled, and the running efficiency of the server is further improved.
Example two
Based on the same inventive concept as the server heat dissipation control method in the foregoing embodiment, as shown in fig. 2, the present application provides a server heat dissipation control system, including:
the configuration module 1 is used for setting multi-stage temperature change grades, carrying out adaptive configuration of a heat dissipation regulation mode and determining a mode regulation mechanism, wherein the heat dissipation regulation mode comprises a conventional regulation mode and an advanced regulation mode, and each stage of temperature change grade comprises a temperature rise rate threshold value and a critical temperature value;
the training module 2 is used for supervising and training an adaptive heat dissipation regulation model by combining the mode regulation mechanism, and the adaptive heat dissipation regulation model is in communication connection with a server heat dissipation system and comprises a parallel conventional regulation branch and an advanced regulation branch;
the activation module 3 is used for reading heat dissipation record data of the radiator and determining a target temperature change grade, and activating a target regulation branch mapped to the target temperature change grade in the adaptive heat dissipation regulation model;
the optimizing module 4 is configured to analyze and determine a heat dissipation regulation and control scheme if the target regulation and control branch is the advanced regulation and control branch, and perform scheme expansion and competitive assimilation optimizing to determine a target heat dissipation scheme, where a heat dissipation system with distributed weights, software optimization and CPU frequency modulation are identified as heat dissipation regulation and control analysis standards;
the management and control module 5 is used for feeding back the target heat dissipation scheme to the server heat dissipation system, performing server heat dissipation management and control based on a radiator, and returning heat dissipation feedback data;
the first judging module 6 is used for judging whether the heat radiation feedback data meet an expected heat radiation standard, if not, a feedback adjustment instruction is generated, and the expected heat radiation standard has a tolerance interval based on technical constraint and scene influence;
and the control analysis module 7 is used for transmitting the feedback adjustment instruction and the heat dissipation feedback data back to the self-adaptive heat dissipation regulation and control model, and repeatedly carrying out heat dissipation control analysis.
Further, the system further comprises:
the first branch determining module is used for determining a conventional heat dissipation flow if the target regulation branch is a conventional regulation branch;
the adjusting module is used for carrying out optional node locking and self-adaptive adjustment on the conventional heat dissipation flow in the conventional regulation branch aiming at the heat dissipation record data, and is used as the target heat dissipation scheme;
and the feedback analysis module is used for feeding the target heat dissipation scheme back to the server heat dissipation system and performing server heat dissipation management and control and feedback analysis.
Further, the system further comprises:
the segmentation module is used for determining a segmentation baseline based on a heat dissipation regulation mode and comprises a reference temperature rise rate and a reference temperature value;
the dividing module is used for dividing the multi-level temperature change grade based on the dividing baseline, determining a first temperature change grade and a second temperature change grade, wherein the multi-level temperature change grade comprises at least three levels and presents an increasing trend;
and the mapping module is used for carrying out mapping configuration of the first temperature change level and the conventional regulation mode, and the second temperature change level and the advanced regulation mode.
Further, the system further comprises:
and the adaptive activation module is used for adaptively activating the conventional regulation branch and the advanced regulation branch along with the heat dissipation regulation demand, and the adaptive heat dissipation regulation model and the server heat dissipation system have data bidirectional interaction.
Further, the system further comprises:
the connection module is used for connecting the advanced regulation branch with an industrial database channel;
the first scheme determining module is used for searching and determining the initialized heat dissipation regulation scheme in the industrial database by taking the heat dissipation record data as an index;
the space construction module is used for determining the degree of freedom of regulation and control based on the heat dissipation system, software optimization and CPU frequency modulation, and constructing diversified optimizing spaces;
and the expansion optimizing module is used for executing expansion optimizing based on the heat dissipation regulation and control scheme in the optimizing space by taking the regulation and control freedom degree as constraint, and determining the target heat dissipation scheme.
Further, the system further comprises:
the random disturbance module is used for carrying out random disturbance of the heat dissipation regulation and control scheme in the optimizing space by taking the regulation and control freedom degree as constraint, and determining an expansion scheme set;
the decrementing module is used for carrying out positive sequence ordering on the expansion scheme set based on scheme superiority and inferiority, determining a preset dividing proportion, dividing the expansion scheme set to determine a plurality of groups of expansion schemes, wherein the preset dividing proportion is sequentially decremented, and the plurality of groups of expansion schemes comprise at least three groups;
the second scheme determining module is used for optimizing and determining the target heat dissipation scheme based on the plurality of groups of expansion schemes;
the inter-group optimization module is used for performing inter-group optimization based on the multiple groups of expansion schemes, and comprises the following steps:
the first optimization module is used for determining an optimal solution of one group of expansion schemes, optimizing the two groups of expansion schemes as an optimization direction and determining a first optimization scheme set;
the second optimization module is used for randomly matching and optimizing the three groups of expansion schemes by taking the group of expansion schemes and the first optimization scheme set as optimization directions, and determining a second optimization scheme set;
and the third scheme determining module is used for sequentially performing iterative optimization, checking the group of expansion schemes, the first optimization scheme set and the second optimization scheme set, and selecting an optimal scheme as the target heat dissipation scheme.
Further, the system further comprises:
the base line setting module is used for setting a heat dissipation base line based on normal operation and control of the server;
the over-frequency regulation analysis module is used for calling a time zone heat dissipation record in a preset time limit, carrying out over-frequency regulation analysis by combining the heat dissipation base line, screening the over-frequency heat dissipation record and calculating the over-frequency duty ratio;
the second judging module is used for judging whether the over-frequency duty ratio meets a threshold value standard or not, and if so, generating a hardware configuration instruction;
and the receiving module is used for executing the hardware optimization processing of the radiator along with the receiving of the hardware configuration instruction.
The foregoing detailed description of the method for controlling server heat dissipation will be apparent to those skilled in the art, and the device disclosed in this embodiment is relatively simple to describe, and the relevant places refer to the method section for description.
The previous description of the disclosed embodiments is provided to enable any person skilled in the art to make or use the present application. Various modifications to these embodiments will be readily apparent to those skilled in the art, and the generic principles defined herein may be applied to other embodiments without departing from the spirit or scope of the application. Thus, the present application is not intended to be limited to the embodiments shown herein but is to be accorded the widest scope consistent with the principles and novel features disclosed herein.
Claims (8)
1. The server heat dissipation control method is characterized by comprising the following steps:
setting multi-stage temperature change grades, carrying out adaptive configuration of a heat dissipation regulation mode, and determining a mode regulation mechanism, wherein the heat dissipation regulation mode comprises a conventional regulation mode and an advanced regulation mode, and each stage of temperature change grade comprises a temperature rise rate threshold value and a critical temperature value;
the mode regulation mechanism is combined, an adaptive heat dissipation regulation model is supervised and trained, and the adaptive heat dissipation regulation model is in communication connection with a server heat dissipation system and comprises a parallel conventional regulation branch and an advanced regulation branch;
reading heat dissipation record data of a radiator, determining a target temperature change grade, and activating a target regulation branch mapped to the target temperature change grade in the self-adaptive heat dissipation regulation model;
if the target regulation branch is the advanced regulation branch, analyzing and determining a heat dissipation regulation scheme, and performing scheme expansion and competitive assimilation optimizing to determine a target heat dissipation scheme, wherein a heat dissipation system with distributed weights is identified, software optimization and CPU frequency modulation are used as heat dissipation regulation analysis standards;
feeding back the target heat dissipation scheme to the server heat dissipation system, performing server heat dissipation management and control based on a radiator, and returning heat dissipation feedback data;
judging whether the heat radiation feedback data meets an expected heat radiation standard, if not, generating a feedback adjustment instruction, wherein the expected heat radiation standard has a tolerant interval based on technical constraint and scene influence;
and the feedback adjustment instruction and the heat dissipation feedback data are returned to the self-adaptive heat dissipation regulation model, and heat dissipation control analysis is repeatedly performed.
2. The method of claim 1, characterized in that the method comprises:
if the target regulation branch is a conventional regulation branch, determining a conventional heat dissipation flow;
for the heat dissipation record data, performing optional node locking and self-adaptive adjustment on the conventional heat dissipation process in the conventional regulation branch to serve as the target heat dissipation scheme;
and feeding the target heat dissipation scheme back to the server heat dissipation system to perform server heat dissipation management and control and feedback analysis.
3. The method of claim 1, wherein the adaptive configuration of the heat dissipation regulation mode is performed by setting a plurality of temperature change levels, the method comprising:
determining a segmentation baseline based on a heat dissipation regulation mode, wherein the segmentation baseline comprises a reference temperature rise rate and a reference temperature value;
classifying and attributing the multi-stage temperature change grades based on the segmentation baseline, determining a first temperature change grade and a second temperature change grade, wherein the multi-stage temperature change grades comprise at least three stages and present an increasing trend;
and performing mapping configuration of the first temperature change level and the conventional regulation mode and mapping configuration of the second temperature change level and the advanced regulation mode.
4. The method of claim 1, wherein the regular regulation branch and the advanced regulation branch are independent operation branches, wherein the adaptive heat dissipation regulation model has data bidirectional interaction with the server heat dissipation system, and wherein the adaptive heat dissipation regulation model is adaptively activated according to heat dissipation regulation requirements.
5. The method of claim 1, wherein the analyzing determines a heat dissipation regulation scheme and performs scheme expansion and competition-equalization optimization, the method comprising:
wherein the advanced regulation branch is connected with an industrial database channel;
searching and determining the initialized heat dissipation regulation scheme in the industrial database by taking the heat dissipation record data as an index;
determining the degree of freedom of regulation and control based on the heat dissipation system, software optimization and CPU frequency modulation, and constructing diversified optimizing space;
and taking the control freedom degree as constraint, executing expansion optimization based on the heat dissipation control scheme in the optimizing space, and determining the target heat dissipation scheme.
6. The method of claim 5, wherein extended optimization based on the heat dissipation regulation scheme is performed in the optimizing space, the method comprising:
taking the regulation and control degrees of freedom as constraint, carrying out random disturbance of the heat dissipation regulation and control scheme in the optimizing space, and determining an expansion scheme set;
the method comprises the steps of carrying out positive sequence ordering of an expansion scheme set based on scheme superiority and inferiority, determining a preset dividing ratio, dividing the expansion scheme set to determine a plurality of groups of expansion schemes, wherein the preset dividing ratio is sequentially decreased, and the plurality of groups of expansion schemes comprise at least three groups;
optimizing and determining the target heat dissipation scheme based on the plurality of groups of expansion schemes;
wherein performing inter-group optimization based on the plurality of groups of expansion schemes comprises:
determining an optimal solution of a group of expansion schemes, taking the optimal solution as an optimization direction, optimizing two groups of expansion schemes, and determining a first optimization scheme set;
taking the group of expansion schemes and the first optimization scheme set as optimization directions, carrying out random matching and optimization on the three groups of expansion schemes, and determining a second optimization scheme set;
and sequentially performing iterative optimization, checking the group of expansion schemes, the first optimization scheme set and the second optimization scheme set, and selecting an optimal scheme as the target heat dissipation scheme.
7. The method of claim 1, characterized in that the method comprises:
setting a heat dissipation base line based on normal operation and control of a server;
calling a time zone heat dissipation record in a preset time limit, performing over-frequency regulation and control analysis by combining the heat dissipation base line, screening the over-frequency heat dissipation record and calculating the over-frequency duty ratio;
judging whether the over-frequency duty ratio meets a threshold value standard or not, and if so, generating a hardware configuration instruction;
and executing the hardware optimization processing of the radiator along with the receiving of the hardware configuration instruction.
8. A server heat dissipation control system, the system comprising:
the configuration module is used for setting multi-stage temperature change grades, carrying out adaptive configuration of a heat dissipation regulation mode and determining a mode regulation mechanism, wherein the heat dissipation regulation mode comprises a conventional regulation mode and an advanced regulation mode, and each stage of temperature change grade comprises a temperature rise rate threshold value and a critical temperature value;
the training module is used for combining the mode regulation and control mechanism, supervising and training an adaptive heat dissipation regulation and control model, and the adaptive heat dissipation regulation and control model is in communication connection with a server heat dissipation system and comprises a parallel conventional regulation and control branch and an advanced regulation and control branch;
the activation module is used for reading heat dissipation record data of the radiator and determining a target temperature change grade, and activating a target regulation branch mapped to the target temperature change grade in the self-adaptive heat dissipation regulation model;
the optimizing module is used for analyzing and determining a heat dissipation regulation and control scheme if the target regulation and control branch is the advanced regulation and control branch, and performing scheme expansion and competitive assimilation optimizing to determine a target heat dissipation scheme, wherein a heat dissipation system with distributed weights is identified, software optimization and CPU frequency modulation are used as heat dissipation regulation and control analysis standards;
the management and control module is used for feeding back the target heat dissipation scheme to the server heat dissipation system, carrying out server heat dissipation management and control based on a radiator and returning heat dissipation feedback data;
the first judging module is used for judging whether the heat radiation feedback data meet expected heat radiation standards, if not, a feedback adjustment instruction is generated, and the expected heat radiation standards have a tolerance interval based on technical constraint and scene influence;
and the control analysis module is used for transmitting the feedback adjustment instruction and the heat dissipation feedback data back to the self-adaptive heat dissipation regulation and control model, and repeatedly carrying out heat dissipation control analysis.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202410150492.6A CN117666741B (en) | 2024-02-02 | 2024-02-02 | Server heat dissipation control method and system |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202410150492.6A CN117666741B (en) | 2024-02-02 | 2024-02-02 | Server heat dissipation control method and system |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN117666741A true CN117666741A (en) | 2024-03-08 |
| CN117666741B CN117666741B (en) | 2024-05-03 |
Family
ID=90073579
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202410150492.6A Active CN117666741B (en) | 2024-02-02 | 2024-02-02 | Server heat dissipation control method and system |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN117666741B (en) |
Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN109375994A (en) * | 2018-09-10 | 2019-02-22 | 西安交通大学 | Data center's task temperature prediction and dispatching method based on RBF neural |
| CN110865666A (en) * | 2019-12-09 | 2020-03-06 | Oppo广东移动通信有限公司 | Temperature control method, temperature control device, storage medium and electronic equipment |
| CN116031777A (en) * | 2023-02-08 | 2023-04-28 | 镇江默勒电器有限公司 | Intelligent control method and system for temperature rise of power distribution cabinet for nuclear power |
| CN116069079A (en) * | 2023-04-06 | 2023-05-05 | 山东海冠电气有限公司 | Intelligent heat dissipation control method and system for intelligent switch cabinet |
-
2024
- 2024-02-02 CN CN202410150492.6A patent/CN117666741B/en active Active
Patent Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN109375994A (en) * | 2018-09-10 | 2019-02-22 | 西安交通大学 | Data center's task temperature prediction and dispatching method based on RBF neural |
| CN110865666A (en) * | 2019-12-09 | 2020-03-06 | Oppo广东移动通信有限公司 | Temperature control method, temperature control device, storage medium and electronic equipment |
| CN116031777A (en) * | 2023-02-08 | 2023-04-28 | 镇江默勒电器有限公司 | Intelligent control method and system for temperature rise of power distribution cabinet for nuclear power |
| CN116069079A (en) * | 2023-04-06 | 2023-05-05 | 山东海冠电气有限公司 | Intelligent heat dissipation control method and system for intelligent switch cabinet |
Also Published As
| Publication number | Publication date |
|---|---|
| CN117666741B (en) | 2024-05-03 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| US9189047B2 (en) | Organizing databases for energy efficiency | |
| US7941426B2 (en) | Optimizing database queries | |
| CN113962142B (en) | Data center temperature prediction method and system based on two-section LSTM | |
| US20100162251A1 (en) | System, method, and computer-readable medium for classifying problem queries to reduce exception processing | |
| US9110949B2 (en) | Generating estimates for query optimization | |
| Zhao et al. | $ k $ NN-DP: handling data skewness in $ kNN $ joins using MapReduce | |
| Zhang et al. | UCPSO: A uniform initialized particle swarm optimization algorithm with cosine inertia weight | |
| CN117193503B (en) | Heat dissipation host power supply operation monitoring method based on Internet of things | |
| US20090281986A1 (en) | Generating Database Query Plans | |
| Zhao | Design and Implementation of an Improved K‐Means Clustering Algorithm | |
| Wang et al. | A hyperparameter optimization algorithm for the LSTM temperature prediction model in data center | |
| US20080147593A1 (en) | Runtime resource sensitive and data driven optimization | |
| Li et al. | Skyline query processing on interval uncertain data | |
| CN117666741B (en) | Server heat dissipation control method and system | |
| Zaefferer et al. | Efficient global optimization with indefinite kernels | |
| Zhou et al. | Data placement in parallel database systems | |
| CN110390011B (en) | Data classification method and device | |
| CN117458261A (en) | Laser packaging system and intelligent heat dissipation method thereof | |
| CN103544191A (en) | Method and device for reading cache data | |
| CN111522420B (en) | Multi-core chip dynamic thermal management method based on power budget | |
| Qi et al. | A Multiple Kernel Learning Model Based on p‐Norm | |
| Zhang et al. | Using a two‐level structure to manage the point location problem in explicit model predictive control | |
| CN112925793B (en) | Distributed hybrid storage method and system for multiple structural data | |
| CN116993315A (en) | Operation and maintenance method and device of cold source equipment, computing equipment and storage medium | |
| US20140089311A1 (en) | System. method, and computer-readable medium for classifying problem queries to reduce exception processing |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |