CN114721488A - Fan rotating speed control method and system - Google Patents
Fan rotating speed control method and system Download PDFInfo
- Publication number
- CN114721488A CN114721488A CN202210268676.3A CN202210268676A CN114721488A CN 114721488 A CN114721488 A CN 114721488A CN 202210268676 A CN202210268676 A CN 202210268676A CN 114721488 A CN114721488 A CN 114721488A
- Authority
- CN
- China
- Prior art keywords
- power consumption
- gpu
- server
- cpu
- fan
- 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
- 238000000034 method Methods 0.000 title claims abstract description 47
- 230000009467 reduction Effects 0.000 claims abstract description 38
- 238000012544 monitoring process Methods 0.000 claims abstract description 12
- 230000008859 change Effects 0.000 claims description 8
- 230000001276 controlling effect Effects 0.000 description 10
- 230000008569 process Effects 0.000 description 9
- 230000006872 improvement Effects 0.000 description 5
- 230000001105 regulatory effect Effects 0.000 description 5
- 230000033228 biological regulation Effects 0.000 description 4
- 238000005457 optimization Methods 0.000 description 4
- 230000002093 peripheral effect Effects 0.000 description 4
- 230000009471 action Effects 0.000 description 3
- 238000013473 artificial intelligence Methods 0.000 description 3
- 238000010586 diagram Methods 0.000 description 3
- 238000012986 modification Methods 0.000 description 3
- 230000004048 modification Effects 0.000 description 3
- 238000005265 energy consumption Methods 0.000 description 2
- 230000017525 heat dissipation Effects 0.000 description 2
- 230000002787 reinforcement Effects 0.000 description 2
- 230000008901 benefit Effects 0.000 description 1
- 230000000694 effects Effects 0.000 description 1
- 238000005516 engineering process Methods 0.000 description 1
- 230000007246 mechanism Effects 0.000 description 1
- 230000000750 progressive effect Effects 0.000 description 1
- 230000001960 triggered effect Effects 0.000 description 1
- 239000002699 waste material Substances 0.000 description 1
Images
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
- G06F1/206—Cooling means comprising thermal management
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F11/00—Error detection; Error correction; Monitoring
- G06F11/30—Monitoring
- G06F11/3058—Monitoring arrangements for monitoring environmental properties or parameters of the computing system or of the computing system component, e.g. monitoring of power, currents, temperature, humidity, position, vibrations
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F11/00—Error detection; Error correction; Monitoring
- G06F11/30—Monitoring
- G06F11/3058—Monitoring arrangements for monitoring environmental properties or parameters of the computing system or of the computing system component, e.g. monitoring of power, currents, temperature, humidity, position, vibrations
- G06F11/3062—Monitoring arrangements for monitoring environmental properties or parameters of the computing system or of the computing system component, e.g. monitoring of power, currents, temperature, humidity, position, vibrations where the monitored property is the power consumption
-
- 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)
- Theoretical Computer Science (AREA)
- Physics & Mathematics (AREA)
- General Engineering & Computer Science (AREA)
- General Physics & Mathematics (AREA)
- Computing Systems (AREA)
- Quality & Reliability (AREA)
- Human Computer Interaction (AREA)
- Cooling Or The Like Of Electrical Apparatus (AREA)
Abstract
The application discloses a fan rotating speed control method and a system, under the condition that a GPU at the current moment and a CPU at the current moment are not subjected to frequency reduction, the total power consumption operation of a monitoring server is executed every time the fan rotating speed reduction operation is executed, the execution of the fan rotating speed reduction operation is stopped until the total power consumption of the server is the lowest within a preset power consumption range, and the total power consumption of the server at least comprises GPU power consumption, CPU power consumption and fan power consumption. By the scheme, the rotating speed of the fan is controlled by combining multi-dimensional data such as the power consumption of the fan, the working frequency of a GPU (graphics processing unit), the working frequency of a CPU (central processing unit), the temperature of the GPU, the temperature of the CPU and the temperature of the CPU, the rotating speed of the fan is controlled in a fine-grained manner, all modules and various application scenes are comprehensively considered in a refined manner, and the rotating speed of the fan is controlled by evaluating the power consumption of the fan and the power consumption of the whole server, so that the whole power consumption of the server is optimized, and the cost of a data center is saved.
Description
Technical Field
The present disclosure relates to the field of fan speed control technologies, and more particularly, to a method and a system for controlling a fan speed.
Background
The fan speed of the current mainstream server is controlled by pulse wave width modulation, the fan control speed is adjusted in real time according to the temperature condition of each module, when the temperature is higher, the fan speed is increased, and when the temperature is lower, the fan speed is reduced, so that the heat dissipation effect is achieved.
The conventional fan rotating speed control is only regulated and controlled from the temperature perspective, the temperature of a CPU (central processing unit), a GPU (graphics processing unit) and other key modules can be usually obtained in real time, the fan rotating speed is increased as long as the temperature of the key modules is higher than a threshold value of the key modules, if the temperature of the key modules is increased, the fan rotating speed is continuously increased, if the temperature of the key modules is reduced, the fan rotating speed is reduced, but the fan rotating speed is generally not set to be very low, and the overall fan runs at a high rotating speed and high power consumption. And the relation between the temperature and the rotating speed is single and fixed, the regulation granularity is coarse, various modules and various application scenes cannot be comprehensively considered in a refined mode, the whole power consumption of the server is not optimized, and the data cost is wasted.
Disclosure of Invention
In view of the above, the application discloses a fan rotation speed control method and system, which aim to control the fan rotation speed in a finer granularity, comprehensively consider each module and various application scenes in a refined manner, and control the fan rotation speed by evaluating the overall power consumption of a server with the power consumption of the fan, so that the overall power consumption of the server is optimized, and the cost of a data center is saved.
In order to achieve the purpose, the technical scheme is as follows:
the first aspect of the present application discloses a method for controlling a rotational speed of a fan, the method comprising:
acquiring the working frequency of a GPU at the current moment, the working frequency of a GPU at the last moment, the working frequency of a CPU at the current moment and the working frequency of a CPU at the last moment in real time;
if the working frequency of the GPU at the current moment is greater than or equal to the working frequency of the GPU at the previous moment, determining that the GPU at the current moment is not subjected to frequency reduction, and if the working frequency of the CPU at the current moment is greater than or equal to the working frequency of the CPU at the previous moment, determining that the CPU at the current moment is not subjected to frequency reduction;
under the condition that neither the GPU at the current moment nor the CPU at the current moment is subjected to frequency reduction, executing the total power consumption operation of the monitoring server every time the operation of reducing the rotating speed of the fan is executed, and stopping executing the operation of reducing the rotating speed of the fan when the total power consumption of the server is the lowest within a preset power consumption range; the total power consumption of the server at least comprises GPU power consumption, CPU power consumption and fan power consumption.
Preferably, the method further comprises the following steps:
if the working frequency of the GPU at the current moment is less than that of the GPU at the previous moment, determining that the GPU at the current moment has frequency reduction;
if the working frequency of the CPU at the current moment is less than that of the CPU at the previous moment, determining that the frequency reduction of the CPU at the current moment exists;
and under the condition that the GPU at the current moment or the CPU at the current moment has frequency reduction, increasing the rotating speed of the fan.
Preferably, the method further comprises the following steps:
acquiring the total power consumption of a server at the current moment and the total power consumption of the server at the previous moment;
if the total power consumption of the server at the current moment is larger than or equal to the total power consumption of the server at the previous moment, determining that the total power consumption of the server at the current moment is not reduced;
and if the total power consumption of the server at the current moment is less than the total power consumption of the server at the last moment, determining that the total power consumption of the server at the current moment is reduced.
Preferably, the method further comprises the following steps:
and if the GPU temperature at the current moment is greater than the preset temperature threshold value, dynamically adjusting the rotating speed of the fan.
Preferably, the method further comprises the following steps:
and if the CPU temperature at the current moment is greater than the preset temperature threshold, dynamically adjusting the rotating speed of the fan.
Preferably, the method further comprises the following steps:
acquiring the working states of the GPUs at different moments;
determining load state change corresponding to the GPU based on the working states of the GPUs at different moments;
and adjusting the rotating speed of the fan in real time through the change of the load state.
A second aspect of the present application discloses a fan speed control system, the system comprising:
the first acquisition unit is used for acquiring the working frequency of the GPU at the current moment, the working frequency of the GPU at the previous moment, the working frequency of the CPU at the current moment and the working frequency of the CPU at the previous moment in real time;
the first determining unit is used for determining that the GPU at the current moment is not subjected to frequency reduction if the working frequency of the GPU at the current moment is greater than or equal to the working frequency of the GPU at the previous moment, and determining that the CPU at the current moment is not subjected to frequency reduction if the working frequency of the CPU at the current moment is greater than or equal to the working frequency of the CPU at the previous moment;
the first execution unit is used for executing the operation of monitoring the total power consumption of the server every time the operation of reducing the rotating speed of the fan is executed under the condition that the GPU at the current moment and the CPU at the current moment are not subjected to frequency reduction, and stopping executing the operation of reducing the rotating speed of the fan until the total power consumption of the server is the lowest within a preset power consumption range; the total power consumption of the server at least comprises GPU power consumption, CPU power consumption and fan power consumption.
Preferably, the method further comprises the following steps:
the second determining unit is used for determining that the frequency reduction of the GPU at the current moment exists if the working frequency of the GPU at the current moment is smaller than the working frequency of the GPU at the previous moment;
a third determining unit, configured to determine that the CPU at the current time has a reduced frequency if the operating frequency of the CPU at the current time is lower than the operating frequency of the CPU at the previous time;
and the second execution unit is used for increasing the rotating speed of the fan under the condition that the frequency reduction exists in the GPU at the current moment or the CPU at the current moment.
Preferably, the method further comprises the following steps:
the second acquisition unit is used for acquiring the total power consumption of the server at the current moment and the total power consumption of the server at the previous moment;
a fourth determining unit, configured to determine that the total power consumption of the server at the current time is not reduced if the total power consumption of the server at the current time is greater than or equal to the total power consumption of the server at the previous time;
a fifth determining unit, configured to determine that the total power consumption of the server at the current time is reduced if the total power consumption of the server at the current time is less than the total power consumption of the server at the previous time.
Preferably, the method further comprises the following steps:
and the first adjusting unit is used for dynamically adjusting the rotating speed of the fan if the GPU temperature at the current moment is greater than a preset temperature threshold value.
According to the technical scheme, the method and the system for controlling the rotating speed of the fan are used for acquiring the working frequency of the GPU at the current moment, the working frequency of the GPU at the previous moment, the working frequency of the CPU at the current moment and the working frequency of the CPU at the previous moment in real time, determining that the GPU at the current moment is not subjected to frequency reduction if the working frequency of the GPU at the current moment is greater than or equal to the working frequency of the GPU at the previous moment, determining that the CPU at the current moment is not subjected to frequency reduction if the working frequency of the CPU at the current moment is greater than or equal to the working frequency of the CPU at the previous moment, and executing the total power consumption operation of the monitoring server every time when the fan rotating speed reducing operation is executed under the condition that the GPU at the current moment and the CPU at the current moment are not subjected to frequency reduction, wherein the total power consumption of the monitoring server at least comprises the power consumption of the GPU, the fan rotating speed reducing operation is stopped when the total power consumption of the server is the lowest within a preset power consumption range, CPU power consumption and fan power consumption. By the scheme, the rotating speed of the fan is controlled by combining the power consumption of the fan, the working frequency of the GPU, the working frequency of the CPU, the temperature of the GPU, the temperature of the CPU and other dimensions, the rotating speed of the fan is controlled in a fine-grained manner, all modules and various application scenes are comprehensively considered in a refined manner, and the rotating speed of the fan is controlled by evaluating the power consumption of the fan and the power consumption of the whole server, so that the whole power consumption of the server is optimized, and the cost of a data center is saved.
Drawings
In order to more clearly illustrate the embodiments of the present application or the technical solutions in the prior art, the drawings needed to be used in the description of the embodiments or the prior art will be briefly introduced below, it is obvious that the drawings in the following description are only embodiments of the present application, and for those skilled in the art, other drawings can be obtained according to the provided drawings without creative efforts.
FIG. 1 is a flow chart illustrating a method for controlling a rotational speed of a fan according to an embodiment of the present disclosure;
FIG. 2 is a schematic diagram illustrating a total power consumption optimization of a server according to an embodiment of the present disclosure;
fig. 3 is a schematic structural diagram of a fan rotational speed control system according to an embodiment of the present disclosure.
Detailed Description
The technical solutions in the embodiments of the present application will be clearly and completely described below with reference to the drawings in the embodiments of the present application, and it is obvious that the described embodiments are only a part of the embodiments of the present application, and not all of the embodiments. All other embodiments, which can be derived by a person skilled in the art from the embodiments given herein without making any creative effort, shall fall within the protection scope of the present application.
In this application, the terms "comprises," "comprising," or any other variation thereof, are intended to cover a non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements does not include only those elements but may include other elements not expressly listed or inherent to such process, method, article, or apparatus. Without further limitation, an element defined by the phrase "comprising an … …" does not exclude the presence of other identical elements in a process, method, article, or apparatus that comprises the element.
Known from the background art, the conventional fan rotating speed control is only regulated and controlled from the temperature angle, the relation between the temperature and the rotating speed is single and fixed, the regulation granularity is thick, various modules and various application scenes cannot be comprehensively considered in a fine mode, the whole power consumption of the server is not optimized, and the data cost is wasted.
In order to solve the above problems, an embodiment of the present application discloses a method and a system for controlling a fan rotation speed, which aim to control the fan rotation speed in a finer granularity, comprehensively consider various modules and various application scenarios in a refined manner, and control the fan rotation speed by evaluating the power consumption of the fan and the power consumption of the whole server, so that the optimization of the whole server power consumption is facilitated, and the cost of a data center is saved. The specific implementation is illustrated by the following examples.
Referring to fig. 1, a flow chart of a method for controlling a fan rotation speed disclosed in an embodiment of the present application is shown, where the method for controlling a fan rotation speed mainly includes the following steps:
s101: the method comprises the steps of obtaining the working frequency of a Graphic Processing Unit (GPU) at the current moment, the working frequency of the GPU at the last moment, the working frequency of a Central Processing Unit (CPU) at the current moment and the working frequency of the CPU at the last moment in real time.
In S101, data such as temperature, power consumption, and operating frequency of each module (such as a GPU and a CPU) is collected in real time by a server Board Management Controller (BMC).
For convenience of understanding the operating frequency of the GPU at the current time and the operating frequency of the GPU at the previous time, and the operating frequency of the CPU at the current time and the operating frequency of the CPU at the previous time, the following description is given by way of example:
for example, if the GPU operating frequency at the current time is 15: 05: 03GPU working frequency, the last time GPU working frequency is 15: 05: 02GPU working frequency.
If the CPU working frequency at the current moment is 13:30: 12CPU working frequency, the CPU working frequency at the last moment is 13:30:11CPU working frequency.
S102: and if the working frequency of the current-time GPU is greater than or equal to the working frequency of the last-time GPU, determining that the current-time GPU does not reduce the frequency, and if the working frequency of the current-time CPU is greater than or equal to the working frequency of the last-time CPU, determining that the current-time CPU does not reduce the frequency.
The benefit of the adjustment is judged by judging whether the performance is lossless (whether GPU is used for reducing frequency and/or whether CPU is used for reducing frequency), and if the performance is certain lossy (whether GPU is used for reducing frequency and/or whether CPU is used for reducing frequency), the rotating speed of the fan is required to be increased if the rotating speed of the fan is not reduced successfully.
Under the conditions that the server is in an idle state, the GPU is in a low load state, and the CPU is in a low load state, the fan is a main power consumption source, the higher the fan power consumption is, the higher the total power consumption of the server is, the lower the fan power consumption is, the lower the total power consumption of the server is, so that the rotating speed of the fan can be reduced, and the total power consumption of the server can be reduced.
And if the total power consumption of the server at the current moment is less than that of the server at the last moment, returning to the step of acquiring the working frequency of the GPU at the current moment, the working frequency of the GPU at the last moment, the working frequency of the CPU at the current moment and the working frequency of the CPU at the last moment in real time.
S103: under the condition that the GPU at the current moment and the CPU at the current moment are not subjected to frequency reduction, executing the total power consumption operation of the monitoring server every time the operation of reducing the rotating speed of the fan is executed, and stopping executing the operation of reducing the rotating speed of the fan when the total power consumption of the server is the lowest within the preset power consumption range; the total power consumption of the server at least comprises GPU power consumption, CPU power consumption and fan power consumption.
In S103, under the condition that neither the GPU at the current time nor the CPU at the current time is down-clocked, the operation of reducing the rotation speed of the fan is executed once and for all, and the operation of monitoring the total power consumption of the server is executed until the lowest total power consumption of the server within the preset power consumption range is obtained, which is an optimization process for obtaining the optimal total power consumption of the server.
The total power consumption of the server is counted by traversing the rotating speed of the fan in real time, and the rotating speed of the fan is regulated and controlled from the perspective of the integral energy consumption ratio of the server, so that the optimal total power consumption of the server is achieved, namely the total power consumption of the server is the lowest within a preset power consumption range.
The total power consumption of the server comprises GPU power consumption, CPU power consumption, fan power consumption, other power consumption and the like, and other power consumption comprises power consumption required by a storage device, an interface, other peripheral devices and related peripheral circuits.
The preset power consumption range is determined by GPU power consumption, CPU power consumption, fan power consumption and other power consumption.
Under the condition that the GPU at the current moment and the CPU at the current moment are not subjected to frequency reduction, the total power consumption operation of the monitoring server is executed every time the fan rotating speed reduction operation is executed, until the total power consumption of the server is the lowest within the preset power consumption range, which is an optimizing process of the total power consumption of the server. In fact, as long as the temperature of the CPU, the GPU and other key modules does not reach the preset temperature threshold, the fan speed may be continuously reduced within a preset range, where the preset range is set by a technician according to an actual situation, and the present application is not specifically limited.
For example, the server is in an idle state, that is, the server does not work, and theoretically, it is feasible that the rotation speed of the fan is almost not rotated, on the premise that modules such as the CPU and the GPU are within a temperature bearing range, and in fact, in many scenes, the server has tasks and sometimes idles, and when this is the case, the regulation and control of the fan are particularly important, and the total power consumption of the server is saved to the maximum extent.
And under the condition that the working efficiency of the server is not influenced, namely the GPU and the CPU cannot exceed the temperature threshold, the rotating speed of the fan is dynamically reduced, and the optimal total power consumption of the server is ensured.
This is a point emphasized in this application, and if the fan rotation speed is not carefully and scientifically regulated and controlled from the perspective of the total power consumption of the server, the energy waste will be brought.
Specifically, from the perspective of the total power consumption of the server, the process of detailed and scientific regulation and control of the fan rotation speed is described by combining the formula (1).
PowerServer＝PowerFAN+PowerGPU+PowerCPU+Powerother(1)
The power server is the total power consumption of the server, the power FAN is the power consumption of the fan, the power GPU is the power consumption of the GPU, the power CPU is the power consumption of the CPU, the power other is other power consumption, and other power consumption comprises the power consumption required by a storage device, an interface, other peripheral equipment and related peripheral circuits.
And judging whether the total power consumption of the server is optimized or not (whether the power consumption of a server system is reduced or not), if so, determining that the total power consumption of the server is optimized, and if not, determining that the total power consumption of the server is not optimized.
And judging the adjustment gain by judging whether the total power consumption of the server is optimized, wherein if the total power consumption of the server is not reduced, the rotating speed of the fan is not reduced successfully, and the rotating speed of the fan needs to be increased.
When the rotating speed of the fan is reduced and the temperature of each module of the server is increased, the total power consumption of the server is naturally increased, and in fact, under the condition that neither the GPU nor the CPU is subjected to frequency reduction, the possible scenes are different, so that a process of optimizing the total power consumption of the server is needed from the perspective of the energy efficiency ratio of the server. The process is described in conjunction with fig. 2, where fig. 2 shows a schematic diagram of the server total power consumption optimization.
In fig. 2, in some working states, the relationship between the fan speed and the total power consumption of the server, and the relationship between the fan speed and the total power consumption of the server are shown, so if the total power consumption of the server is to be found to be optimal, it is a process that the total power consumption of the server needs to be optimized. In fig. 2, the ordinate (1600, 1700, 1800, 1900, 2000, 2100) and y are the total power consumption of the server, and the abscissa (100%, 90%, 80%, 70%, 60%, 50%) and x are the fan speed.
For example, the GPU is kept in a certain working state, then the rotation speed of the fan is continuously traversed in real time, and the total power consumption of the server is counted, so that the rotation speed of the fan is regulated and controlled from the perspective of the overall energy consumption ratio of the server, and the optimal total power consumption of the server is achieved.
The working states of the GPUs at different moments are obtained, the load state change corresponding to the GPU is determined based on the working states of the GPUs at the different moments, and the rotating speed of the fan is adjusted in real time through the load state change.
The GPU is provided with a plurality of operation units, the number of the operation units in the GPU is used for determining the working state of the GPU, the load degree of the GPU is determined according to the working state of the GPU, when the operation units in the GPU are operated, the heat productivity of the GPU is high, the load of the GPU is high, the heat quantity generated by the GPU is high, the total power consumption of a server is increased, and therefore the rotating speed of a fan can be dynamically increased.
Fig. 2 also shows that it is not a simple matter to consider the lowest fan speed, but rather the optimal total power consumption of the server. But the power consumption of the GPU, the power consumption of the CPU and the like need to be detected in real time, then comparison is carried out, the rotating speed of the fan is adjusted and the like, and the optimal total power consumption of the server is achieved.
The server BMC takes the collected temperature and power consumption information of each module as input, guarantees no performance loss and normal working temperature of each module as premise, takes the lowest power consumption of the server as output, takes an AI reinforced learning algorithm as execution logic, and finds out the optimal fan rotating speed at a reasonable temperature according to load conditions under different working scenes.
The method comprises the steps of recording the quality of a set result every time through an AI (artificial intelligence) reinforcement learning algorithm, further carrying out self judgment and improvement, and finally finding out the optimal rotating speed of the fan quickly and accurately according to status at any moment after mass data reinforcement and improvement.
In the control of the rotating speed of the fan, the power consumption of the fan is important, and the total power consumption of the server is the power consumption of the fan, the power consumption of a GPU, the power consumption of a CPU and other power consumption. Theoretically, the rotating speed of the fan is set to be the maximum, both the GPU and the CPU can work at normal temperature, but the total power consumption of the server is too large when the rotating speed of the fan is set to be higher by one gear; if the fan rotation speed is adjusted too low, the heat dissipation capacity of the GPU and the CPU is insufficient, the GPU and the CPU may touch respective upper temperature limits, the protection mechanisms of the GPU and the CPU are triggered, and therefore the frequency is reduced, the working efficiency is reduced, and the method is not preferable.
For example, if the fan speed is at full speed, 650W is consumed; 90% speed, 500W consumption; 80% speed, consume 350W; 70% speed, 250W consumed; and the normal operation of the whole server consumes more than 2000W. Therefore, the rotating speed of the fan is reduced as much as possible, the GPU and the CPU are enabled to be close to respective critical values of the upper temperature limits, the working efficiency is not affected, meanwhile, the power consumption of the fan is omitted, and the total power consumption of the server is enabled to be the lowest.
Different fan rotating speeds are set for the actual working scene of the server, result judgment is carried out by monitoring information such as temperature, power consumption and working frequency, subsequent setting strategies are guided, and finally the optimal rotating speed of the fan can be quickly and accurately found according to the working state of the server at any moment after intensive learning and improvement of mass data.
In the embodiment of the application, the fan rotating speed is controlled by combining multidimensional data such as the fan power consumption, the GPU working frequency, the CPU working frequency, the GPU temperature, the CPU temperature and the like, the fan rotating speed is controlled in a fine-grained manner, all modules and various application scenes are comprehensively considered in a fine-grained manner, and the fan rotating speed is controlled by evaluating the power consumption of the whole server, so that the whole power consumption of the server is optimized, and the cost of a data center is saved.
Based on the method for controlling the fan speed disclosed in fig. 1 in the foregoing embodiment, an embodiment of the present application also discloses a system for controlling the fan speed, and as shown in fig. 3, the system for controlling the fan speed mainly includes a first obtaining unit 301, a first determining unit 302, and a first executing unit 303.
The first obtaining unit 301 is configured to obtain, in real time, an operating frequency of the GPU at the current time, an operating frequency of the GPU at the previous time, an operating frequency of the CPU at the current time, and an operating frequency of the CPU at the previous time.
The first determining unit 302 is configured to determine that the GPU at the current time is not down-converted if the operating frequency of the GPU at the current time is greater than or equal to the operating frequency of the GPU at the previous time, and determine that the CPU at the current time is not down-converted if the operating frequency of the CPU at the current time is greater than or equal to the operating frequency of the CPU at the previous time.
A first executing unit 303, configured to execute a total power consumption operation of the monitoring server every time a fan rotation speed reduction operation is executed under the condition that neither the GPU at the current time nor the CPU at the current time is subjected to frequency reduction, and stop executing the fan rotation speed reduction operation until the total power consumption of the server is the lowest within a preset power consumption range; the total power consumption of the server at least comprises GPU power consumption, CPU power consumption and fan power consumption.
Further, the fan speed control system further comprises a second determining unit, a third determining unit and a second executing unit.
And the second determining unit is used for determining that the frequency reduction of the GPU at the current moment exists if the working frequency of the GPU at the current moment is smaller than that of the GPU at the previous moment.
And the third determining unit is used for determining that the frequency reduction exists in the CPU at the current moment if the working frequency of the CPU at the current moment is less than that of the CPU at the previous moment.
And the second execution unit is used for increasing the rotating speed of the fan under the condition that the frequency reduction exists in the GPU at the current moment or the CPU at the current moment.
Further, the fan speed control system further comprises a second obtaining unit, a fourth determining unit and a fifth determining unit.
And the second acquisition unit is used for acquiring the total power consumption of the server at the current moment and the total power consumption of the server at the previous moment.
And the fourth determining unit is used for determining that the total power consumption of the server at the current moment is not reduced if the total power consumption of the server at the current moment is greater than or equal to the total power consumption of the server at the previous moment.
A fifth determining unit, configured to determine that the total power consumption of the server at the current time is reduced if the total power consumption of the server at the current time is smaller than the total power consumption of the server at the previous time.
Further, the fan speed control system also comprises a first adjusting unit.
And the first adjusting unit is used for dynamically adjusting the rotating speed of the fan if the GPU temperature at the current moment is greater than a preset temperature threshold value.
Further, the fan speed control system further comprises a second adjusting unit.
And the second adjusting unit is used for dynamically adjusting the rotating speed of the fan if the temperature of the CPU at the current moment is greater than a preset temperature threshold value.
Further, the fan rotating speed control system further comprises a third obtaining unit, a sixth determining unit and a second adjusting unit.
And the third acquisition unit is used for acquiring the working states of the GPUs at different moments.
And the sixth determining unit is used for determining the load state change corresponding to the GPU based on the working states of the GPUs at different moments.
And the second adjusting unit adjusts the rotating speed of the fan in real time through the change of the load state.
In the embodiment of the application, the rotating speed of the fan is controlled by combining multidimensional data such as the power consumption of the fan, the working frequency of a GPU, the working frequency of a CPU, the temperature of the GPU, the temperature of the CPU and the like, the rotating speed of the fan is controlled in a fine-grained manner, various modules and various application scenes are comprehensively considered in a fine-grained manner, and the rotating speed of the fan is controlled by evaluating the power consumption of the fan and the power consumption of the whole server, so that the whole power consumption of the server is optimized, and the cost of a data center is saved.
While, for purposes of simplicity of explanation, the foregoing method embodiments are presented as a series of acts or combinations, it will be appreciated by those of ordinary skill in the art that the present application is not limited by the illustrated ordering of acts, as some steps may occur in other orders or concurrently depending on the application. Further, those skilled in the art will recognize that the embodiments described in this specification are preferred embodiments and that acts or modules referred to are not necessarily required for this application.
It should be noted that, in the present specification, the embodiments are all described in a progressive manner, each embodiment focuses on differences from other embodiments, and the same and similar parts among the embodiments may be referred to each other. For the system-class embodiment, since it is basically similar to the method embodiment, the description is simple, and for the relevant points, reference may be made to the partial description of the method embodiment.
The steps in the method of each embodiment of the present application may be sequentially adjusted, combined, and deleted according to actual needs.
Finally, it should also be noted that, herein, relational terms such as first and second, and the like may be used solely to distinguish one entity or action from another entity or action without necessarily requiring or implying any actual such relationship or order between such entities or actions.
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 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.
The foregoing is only a preferred embodiment of the present application and it should be noted that those skilled in the art can make several improvements and modifications without departing from the principle of the present application, and these improvements and modifications should also be considered as the protection scope of the present application.
Claims (10)
1. A method of controlling fan speed, the method comprising:
acquiring the working frequency of a GPU at the current moment, the working frequency of a GPU at the last moment, the working frequency of a CPU at the current moment and the working frequency of a CPU at the last moment in real time;
if the working frequency of the GPU at the current moment is greater than or equal to the working frequency of the GPU at the previous moment, determining that the GPU at the current moment is not subjected to frequency reduction, and if the working frequency of the CPU at the current moment is greater than or equal to the working frequency of the CPU at the previous moment, determining that the CPU at the current moment is not subjected to frequency reduction;
under the condition that neither the GPU at the current moment nor the CPU at the current moment is subjected to frequency reduction, executing the total power consumption operation of the monitoring server every time the operation of reducing the rotating speed of the fan is executed, and stopping executing the operation of reducing the rotating speed of the fan when the total power consumption of the server is the lowest within a preset power consumption range; the total power consumption of the server at least comprises GPU power consumption, CPU power consumption and fan power consumption.
2. The method of claim 1, further comprising:
if the working frequency of the GPU at the current moment is less than that of the GPU at the previous moment, determining that the GPU at the current moment has frequency reduction;
if the working frequency of the CPU at the current moment is less than that of the CPU at the previous moment, determining that the frequency reduction of the CPU at the current moment exists;
and under the condition that the GPU at the current moment or the CPU at the current moment has frequency reduction, increasing the rotating speed of the fan.
3. The method of claim 1, further comprising:
acquiring the total power consumption of a server at the current moment and the total power consumption of the server at the last moment;
if the total power consumption of the server at the current moment is larger than or equal to the total power consumption of the server at the previous moment, determining that the total power consumption of the server at the current moment is not reduced;
and if the total power consumption of the server at the current moment is less than the total power consumption of the server at the last moment, determining that the total power consumption of the server at the current moment is reduced.
4. The method of claim 1, further comprising:
and if the GPU temperature at the current moment is greater than the preset temperature threshold value, dynamically adjusting the rotating speed of the fan.
5. The method of claim 1, further comprising:
and if the CPU temperature at the current moment is greater than the preset temperature threshold, dynamically adjusting the rotating speed of the fan.
6. The method of claim 1, further comprising:
acquiring the working states of the GPUs at different moments;
determining load state change corresponding to the GPU based on the working states of the GPUs at different moments;
and adjusting the rotating speed of the fan in real time through the change of the load state.
7. A fan speed control system, the system comprising:
the first acquisition unit is used for acquiring the working frequency of the GPU at the current moment, the working frequency of the GPU at the previous moment, the working frequency of the CPU at the current moment and the working frequency of the CPU at the previous moment in real time;
the first determining unit is used for determining that the GPU at the current moment is not subjected to frequency reduction if the working frequency of the GPU at the current moment is greater than or equal to the working frequency of the GPU at the previous moment, and determining that the CPU at the current moment is not subjected to frequency reduction if the working frequency of the CPU at the current moment is greater than or equal to the working frequency of the CPU at the previous moment;
the first execution unit is used for executing the operation of monitoring the total power consumption of the server every time the operation of reducing the rotating speed of the fan is executed under the condition that the GPU at the current moment and the CPU at the current moment are not subjected to frequency reduction, and stopping executing the operation of reducing the rotating speed of the fan until the total power consumption of the server is the lowest within a preset power consumption range; the total power consumption of the server at least comprises GPU power consumption, CPU power consumption and fan power consumption.
8. The system of claim 7, further comprising:
the second determining unit is used for determining that the frequency reduction of the GPU at the current moment exists if the working frequency of the GPU at the current moment is smaller than the working frequency of the GPU at the previous moment;
a third determining unit, configured to determine that the CPU at the current time has a reduced frequency if the operating frequency of the CPU at the current time is lower than the operating frequency of the CPU at the previous time;
and the second execution unit is used for increasing the rotating speed of the fan under the condition that the frequency reduction exists in the GPU or the CPU at the current moment.
9. The system of claim 7, further comprising:
the second acquisition unit is used for acquiring the total power consumption of the server at the current moment and the total power consumption of the server at the previous moment;
a fourth determining unit, configured to determine that the total power consumption of the server at the current time is not reduced if the total power consumption of the server at the current time is greater than or equal to the total power consumption of the server at the previous time;
a fifth determining unit, configured to determine that the total power consumption of the server at the current time is reduced if the total power consumption of the server at the current time is smaller than the total power consumption of the server at the previous time.
10. The system of claim 7, further comprising:
and the first adjusting unit is used for dynamically adjusting the rotating speed of the fan if the GPU temperature at the current moment is greater than a preset temperature threshold value.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202210268676.3A CN114721488B (en) | 2022-03-18 | 2022-03-18 | Fan rotating speed control method and system |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202210268676.3A CN114721488B (en) | 2022-03-18 | 2022-03-18 | Fan rotating speed control method and system |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN114721488A true CN114721488A (en) | 2022-07-08 |
| CN114721488B CN114721488B (en) | 2024-07-19 |
Family
ID=82236810
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202210268676.3A Active CN114721488B (en) | 2022-03-18 | 2022-03-18 | Fan rotating speed control method and system |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN114721488B (en) |
Cited By (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN116643952A (en) * | 2023-07-27 | 2023-08-25 | 安擎计算机信息股份有限公司 | Server power consumption management method and device |
| CN116755982A (en) * | 2023-05-15 | 2023-09-15 | 合芯科技(苏州)有限公司 | Method and system for calculating energy consumption ratio of server |
| CN116755542A (en) * | 2023-06-16 | 2023-09-15 | 北京市合芯数字科技有限公司 | Whole machine power consumption reduction method, system, substrate management controller and server |
Citations (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110206747A (en) * | 2019-04-16 | 2019-09-06 | 苏州浪潮智能科技有限公司 | A kind of server fan regulation method and system based on component power consumption |
| US20200192444A1 (en) * | 2017-08-30 | 2020-06-18 | New H3C Information Technologies Co., Ltd. | Controlling fan speed of server |
-
2022
- 2022-03-18 CN CN202210268676.3A patent/CN114721488B/en active Active
Patent Citations (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20200192444A1 (en) * | 2017-08-30 | 2020-06-18 | New H3C Information Technologies Co., Ltd. | Controlling fan speed of server |
| CN110206747A (en) * | 2019-04-16 | 2019-09-06 | 苏州浪潮智能科技有限公司 | A kind of server fan regulation method and system based on component power consumption |
Cited By (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN116755982A (en) * | 2023-05-15 | 2023-09-15 | 合芯科技(苏州)有限公司 | Method and system for calculating energy consumption ratio of server |
| CN116755542A (en) * | 2023-06-16 | 2023-09-15 | 北京市合芯数字科技有限公司 | Whole machine power consumption reduction method, system, substrate management controller and server |
| CN116755542B (en) * | 2023-06-16 | 2024-06-04 | 北京市合芯数字科技有限公司 | Whole machine power consumption reduction method, system, substrate management controller and server |
| CN116643952A (en) * | 2023-07-27 | 2023-08-25 | 安擎计算机信息股份有限公司 | Server power consumption management method and device |
Also Published As
| Publication number | Publication date |
|---|---|
| CN114721488B (en) | 2024-07-19 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN114721488A (en) | Fan rotating speed control method and system | |
| DE102007048505B4 (en) | Server configured to manage performance and performance | |
| CN102232205B (en) | Forced idle of a data processing system | |
| Bansal et al. | Dynamic speed scaling to manage energy and temperature | |
| CN103593041B (en) | The forced idle of data handling system | |
| DE112006003444B4 (en) | Method and apparatus for detecting processor state transitions | |
| DE102009058426B4 (en) | Method and device for power management of a processor | |
| CN1251042C (en) | Bus clock controlling device and method for portable computer | |
| CN101379453B (en) | Method and apparatus for using dynamic workload characteristics to control CPU frequency and voltage scaling | |
| US8001403B2 (en) | Data center power management utilizing a power policy and a load factor | |
| US20030023885A1 (en) | Automated power management system for a network of computers | |
| US20070011480A1 (en) | Processor power management | |
| DE112013005278T5 (en) | Dynamically balancing performance across a plurality of processor domains in accordance with performance balancing BIAS | |
| CN113672383A (en) | Cloud computing resource scheduling method, system, terminal and storage medium | |
| CN101477398A (en) | Terminal power consumption control method and apparatus | |
| US9652027B2 (en) | Thread scheduling based on performance state and idle state of processing units | |
| CN105677000A (en) | Dynamic voltage and frequency scaling system and method | |
| CN113535409A (en) | Server-free computing resource distribution system oriented to energy consumption optimization | |
| CN115963914A (en) | Hybrid strategy-based intelligent mobile equipment power supply management method and device | |
| US6873136B2 (en) | Smart VRM to extend the battery life | |
| CN113568805B (en) | Real-time adjustment method and device for power consumption of server and electronic equipment | |
| DE112007003007T5 (en) | Method and apparatus for power management in a processor | |
| JPH10187636A (en) | Power-consumption reducing cluster system | |
| CN112394803B (en) | Domestic server power consumption real-time dynamic adjusting method | |
| EP4359882A1 (en) | System and method for controlling electrical current supply in a multi-processor core system via instruction per cycle reduction |
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 | ||
| CB02 | Change of applicant information | ||
| CB02 | Change of applicant information |
Country or region after: China Address after: 200120 room a-522, 188 Yesheng Road, Lingang xinpian District, China (Shanghai) pilot Free Trade Zone, Pudong New Area, Shanghai Applicant after: Shanghai Suiyuan Technology Co.,Ltd. Address before: 200120 room a-522, 188 Yesheng Road, Lingang xinpian District, China (Shanghai) pilot Free Trade Zone, Pudong New Area, Shanghai Applicant before: SHANGHAI ENFLAME TECHNOLOGY Co.,Ltd. Country or region before: China |
|
| GR01 | Patent grant | ||
| GR01 | Patent grant |