CN103062091B - A kind of fan intelligent control method - Google Patents
A kind of fan intelligent control method Download PDFInfo
- Publication number
- CN103062091B CN103062091B CN201310030987.7A CN201310030987A CN103062091B CN 103062091 B CN103062091 B CN 103062091B CN 201310030987 A CN201310030987 A CN 201310030987A CN 103062091 B CN103062091 B CN 103062091B
- Authority
- CN
- China
- Prior art keywords
- node
- fan
- temperature
- pwm
- regulation
- 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.)
- Expired - Fee Related
Links
- 238000000034 method Methods 0.000 title claims abstract description 21
- 230000033228 biological regulation Effects 0.000 claims abstract description 33
- 238000001914 filtration Methods 0.000 claims description 2
- 230000001276 controlling effect Effects 0.000 description 13
- 230000005855 radiation Effects 0.000 description 8
- 238000005516 engineering process Methods 0.000 description 2
- 230000004044 response Effects 0.000 description 2
- 239000002699 waste material Substances 0.000 description 2
- 230000001133 acceleration Effects 0.000 description 1
- 230000009286 beneficial effect Effects 0.000 description 1
- 230000005540 biological transmission Effects 0.000 description 1
- 238000004891 communication Methods 0.000 description 1
- 238000001816 cooling Methods 0.000 description 1
- 230000007812 deficiency Effects 0.000 description 1
- 238000012938 design process Methods 0.000 description 1
- 230000017525 heat dissipation Effects 0.000 description 1
- 238000007726 management method Methods 0.000 description 1
- 239000000203 mixture Substances 0.000 description 1
- 230000002093 peripheral effect Effects 0.000 description 1
- 230000001105 regulatory effect Effects 0.000 description 1
- 230000000630 rising effect Effects 0.000 description 1
- 230000007704 transition Effects 0.000 description 1
Abstract
The invention provides a kind of fan intelligent control method, its concrete regulation process is: first obtain the height dimension size of single fan structure, and according to the height dimension size of this fan structure, server node is carried out to region division; Obtain the temperature conditions of server node in grouping; Each node is according to the pwm signal of said temperature situation output regulation and control rotation speed of the fan; Relatively judge the dutycycle size of PWM; Switch the large actual fan control of conduct of PWM dutycycle. Compared to the prior art this one fan intelligent control method, has ensured high efficiency, the Low Noise Design of multiserver node regulation and control, realizes the intellectuality of fan regulation and controlling, has higher reliability, stability.
Description
Technical field
The present invention relates to computer communication technology field, specifically a kind of high efficiency, low noise fan intelligent control method that ensures the regulation and control of multiserver node.
Background technology
Current server system, the topology that adopts cluster multinode to distribute is more and more general, be that multiple server nodes gather together, set up server cluster, server cluster processor active task requires more and more higher, and it is more and more faster to outside heat radiation responding ability, in completing big data quantity high performance computing, the higher demand that peripheral fan regulation and controlling design is proposed, fan regulation and controlling need to be taken into account real-time response ability, power consumption size, noise control; The real-time response ability of fan regulation and controlling requires in the time that server node temperature raises fast, can control fan steadily speed-raising as early as possible, effectively suppresses the rising of server node temperature; In fan regulation and controlling, power consumption size is controlled, and is to require to reduce at server node processor active task, after node temperature declines, fan may operate at the state of the slow-speed of revolution, low-power consumption, noise control in fan regulation and controlling is the acceleration and deceleration that requirement fan is level and smooth, avoids the speed quantitative change that increases, and produces noise. In the group system of multiserver node, fan radiating system is shared by Servers-all node, and how fan cooling carries out Effective Regulation in conjunction with the temperature information of Servers-all node, becomes a difficult problem that needs solution in current multiserver node system; Because operand, the temperature of each server node are all different, the all different wind speed requirements of correspondence of each node, if adopt the equal controlling fan regulation and control of each server single node, the regulation and control of fan can enter unordered mix, cannot form effective regulation and control. Today, server cluster overall power is larger, in this case, ensures high efficiency, the intelligent design of the heat radiation of server multinode group system, has great importance for the stability of server multi-node system.
Currently can only take into account entirety or constant mode to the heat dissipation design of server multi-node system, include limit of consideration in by all nodes, overall power hour, rotation speed of the fan reduces, or fan to be fixed on 80% high invariablenes turning speed constant; Same fan provides heat radiation air quantity for multiple server nodes, the unbalanced poor efficiency that causes unified heat radiation of server operation load capacity; Heat radiation effectively in time when taking into account overall heat radiation strategy and cannot ensureing that single node carrying capacity is larger, the constant mode strategy that dispels the heat cannot dynamically change wind speed in real time, easily causes the waste of the energy, and the noise under high rotating speed also cannot be effectively controlled simultaneously. In order to ensure the stable operation of multi-node server system, in system running, stablize the design of effective fan regulation and controlling particularly important, and become and determine one of server system stability key element.
For the problems referred to above that run in current fan regulation and controlling design process, need a kind of fan intelligent control method for designing based on multiserver node.
Summary of the invention
Technical assignment of the present invention is to solve the deficiencies in the prior art, and a kind of fan intelligent control method based on multiserver node is provided.
Technical scheme of the present invention realizes in the following manner, this one fan intelligent control method, and its concrete regulation process is:
Step 1, first obtain the height dimension size of single fan structure, and according to the height dimension size of this fan structure, server node is carried out to region division;
Step 2, obtain the temperature conditions of server node in grouping;
Step 3, each node are according to the pwm signal of said temperature situation output regulation and control rotation speed of the fan;
Step 4, relatively judge the dutycycle size of PWM;
Step 5, the large actual fan control of conduct of switching PWM dutycycle.
In described step 2, the temperature of node is obtained by the temperature sensor collection being positioned on mainboard, gathers the temperature value of CPU, internal memory, south bridge annex, gets after average, as the temperature value of this node.
The detailed process of described step 4 is: by the PWM waveform signal of each server node output, circuit after filtering, obtain corresponding analog voltage, and compare the amplitude size of all analog quantitys, the PWM dutycycle maximum of the large correspondence of amplitude of analog quantity, and then the temperature of corresponding server node is the highest.
Described PWM filter circuit adopts RC capacitance-resistance filter, and resistance value is taken as 4.7 kilo-ohms, and the value of electric capacity is taken as 470uf.
The detailed process of described step 5 is: judge in node after the high node of temperature, can, using the PWM waveform signal of node output the highest this temperature as to the speed controling signal that should organize fan, complete fan control.
The beneficial effect that the present invention compared with prior art produced is:
A kind of fan intelligent control method of the present invention solves current in multiserver node system running, fan regulation and controlling design easily causes the waste of the energy, the problem that noise under high rotating speed also cannot be effectively controlled simultaneously, high efficiency, the Low Noise Design of the regulation and control of multiserver node are ensured, realize the intellectuality of fan regulation and controlling, there is higher reliability, stability, have great importance for the radiation stability of server system.
Brief description of the drawings
Accompanying drawing 1 is regulation and control schematic flow sheet of the present invention.
Detailed description of the invention
Below in conjunction with accompanying drawing, a kind of fan intelligent control method of the present invention is described in detail below.
The present invention is with the control theory strong point, specifically utilizes a kind of Power Management Design method of high efficiency, low ripple, solves currently in server master board running, and the delivery efficiency of power supply and ripple cannot reach the problem of taking into account. , taking three server nodes as example, as shown in Figure 1, now provide a kind of fan intelligent control method now, its concrete regulation process is:
Step 1, first obtain the height dimension size of single fan structure, and according to the height dimension size of this fan structure, server node is carried out to region division, the height dimension of general single fan is 12cm, server thickness is 4.45cm (1U), therefore using three server nodes (corresponding 3U height) as one group; Above-mentioned this division belongs to minimum fan regulation and controlling region, can adopt unified regulating strategy by row's transverse fan, and this dividing mode can realize the grouping of whole server cluster, subregion optimizing thermal solution.
Step 2, obtain the temperature conditions of server node in grouping, the temperature of each node is collected and is obtained by being positioned at multiple temperature sensors on mainboard, mainly gathers the temperature value of CPU, internal memory, south bridge annex, gets after average, as the temperature value of this node.
Step 3, each node be according to the pwm signal of said temperature situation output regulation and control rotation speed of the fan, the pwm signal of output regulation and control rotation speed of the fan, and node 1 exports that PWM1, node 2 export PWM2, node 3 is exported PWM3. This pwm signal is exported by the managing chip AST2300 of mainboard.
Step 4, PWM1, PWM2, PWM3 waveform signal that server node is exported, through RC filter circuit, obtain corresponding analog voltage, and compare the amplitude size of three analog quantitys through comparison amplifier LM324, the PWM dutycycle of the large correspondence of amplitude of analog quantity is also just large, and then the temperature of corresponding server node is also just high; Judge in three nodes after the high node of temperature, can, using the PWM waveform signal of node output the highest this temperature as to the speed controling signal that should organize fan, reach the optimum fan control of grouping.
Resistance R value in described filter circuit is taken as 4.7 kilo-ohms, the value of electric capacity is taken as 470uf, it is to play to maximize level and smooth PWM waveform that two parameter values are got this value object one, the 2nd, the time that discharges and recharges that can strengthen resistance, avoid small disturbance, cause circuit frequently to switch, improve the stability of the regulation and control judgement of circuit,, make efficiency of transmission and ripple reach optimum
Step 5, the large actual fan control of conduct of switching PWM dutycycle.
For avoid pwm signal follow node change switch in, the fan noise bringing, the PWM waveform signal of the highest node of temperature output is as to before the speed controling signal that should organize fan, all compare with current use value, if deviation is larger, the mode that adopts dutycycle progressively to approach, be that PWM dutycycle is through a linear regulation region, reach the slow slow optimum fan noise control of falling that rises, there is dutycycle A in dutycycle B change procedure, adopt the mode of linear interpolation to increase progressively uniformly, ensure smooth transition.
Through detailed above enforcement, we can carry out the design of fan regulation and controlling very easily, have not only reached the requirement of fan regulation and controlling high efficiency, and realize the low noise requirement of fan, improve regulation and control efficiency and accuracy, improved the heat radiation reliability and stability of server system.
Claims (2)
1. a fan intelligent control method, is characterized in that: its concrete regulation process is:
Step 1, first obtain the height dimension size of single fan structure, and according to the height dimension size of this fan structure, server node is carried out to region division;
Step 2, obtain the temperature conditions of server node in grouping, the temperature of this node is collected and is obtained by being positioned at temperature sensor on mainboard, gathers the temperature value of CPU, internal memory, south bridge annex, gets after average, as the temperature value of this node;
Step 3, each node are according to the pwm signal of said temperature situation output regulation and control rotation speed of the fan;
Step 4, relatively judge the dutycycle size of PWM, by the PWM waveform signal of the each node output of server, circuit after filtering, obtain corresponding analog voltage, and compare the amplitude size of all analog voltages, the maximum corresponding PWM dutycycle maximum of amplitude of analog voltage, and then the temperature of corresponding server node is the highest;
Step 5, the large actual fan control of conduct of switching PWM dutycycle, judge in node after the high node of temperature, the rotation speed of the fan control signal of the PWM waveform signal that node the highest this temperature can be exported in this node region of living in, completes fan control.
2. a kind of fan intelligent control method according to claim 1, is characterized in that: described filter circuit adopts RC capacitance-resistance filter, and resistance value is taken as 4.7 kilo-ohms, and the value of electric capacity is taken as 470uf.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201310030987.7A CN103062091B (en) | 2013-01-28 | 2013-01-28 | A kind of fan intelligent control method |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201310030987.7A CN103062091B (en) | 2013-01-28 | 2013-01-28 | A kind of fan intelligent control method |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN103062091A CN103062091A (en) | 2013-04-24 |
| CN103062091B true CN103062091B (en) | 2016-06-08 |
Family
ID=48104895
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201310030987.7A Expired - Fee Related CN103062091B (en) | 2013-01-28 | 2013-01-28 | A kind of fan intelligent control method |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN103062091B (en) |
Families Citing this family (9)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN104750212A (en) * | 2013-12-31 | 2015-07-01 | 鸿富锦精密电子(天津)有限公司 | Fan control system |
| CN104391555B (en) * | 2014-12-05 | 2017-08-25 | 浪潮集团有限公司 | A kind of server fan speed regulating method |
| CN104454616A (en) * | 2014-12-26 | 2015-03-25 | 浪潮电子信息产业股份有限公司 | Method for accurately controlling rotating speed of fan |
| US9785134B2 (en) * | 2015-03-09 | 2017-10-10 | Quanta Computer, Inc. | Fan duty control for multi-node computing device |
| CN107045379B (en) * | 2016-02-05 | 2021-01-29 | 华为技术有限公司 | Cabinet server temperature control method and device |
| CN106774745A (en) * | 2016-12-27 | 2017-05-31 | 郑州云海信息技术有限公司 | Server fan rotating property test rotation speed of the fan changing method and control system |
| CN106802844A (en) * | 2016-12-30 | 2017-06-06 | 郑州云海信息技术有限公司 | A kind of fan control system and method for fan control signal state |
| CN106678067B (en) * | 2017-03-10 | 2018-02-23 | 郑州云海信息技术有限公司 | Control method for fan and system in a kind of multi-partition server system |
| CN113653666B (en) * | 2021-09-28 | 2023-05-16 | 四川虹美智能科技有限公司 | PG fan control device |
Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101008855A (en) * | 2006-01-25 | 2007-08-01 | 英业达股份有限公司 | System and method for controlling rotational speed of fan |
| CN101761500A (en) * | 2008-12-24 | 2010-06-30 | 大唐移动通信设备有限公司 | Method and device for controlling rotating speed of fan and method and device for controlling heat dissipation speed |
| CN102307447A (en) * | 2011-08-29 | 2012-01-04 | 中兴通讯股份有限公司 | Heat dissipation method and device |
| CN102520774A (en) * | 2011-11-22 | 2012-06-27 | 浪潮(北京)电子信息产业有限公司 | Radiating system of blade server and radiating control method |
| CN102740655A (en) * | 2011-04-07 | 2012-10-17 | 鸿富锦精密工业（深圳）有限公司 | Data center and heat radiation control system thereof |
Family Cites Families (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20080281475A1 (en) * | 2007-05-09 | 2008-11-13 | Tyan Computer Corporation | Fan control scheme |
| US8292593B2 (en) * | 2009-08-28 | 2012-10-23 | Oracle America, Inc. | System for minimizing mechanical and acoustical fan noise coupling |
-
2013
- 2013-01-28 CN CN201310030987.7A patent/CN103062091B/en not_active Expired - Fee Related
Patent Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101008855A (en) * | 2006-01-25 | 2007-08-01 | 英业达股份有限公司 | System and method for controlling rotational speed of fan |
| CN101761500A (en) * | 2008-12-24 | 2010-06-30 | 大唐移动通信设备有限公司 | Method and device for controlling rotating speed of fan and method and device for controlling heat dissipation speed |
| CN102740655A (en) * | 2011-04-07 | 2012-10-17 | 鸿富锦精密工业（深圳）有限公司 | Data center and heat radiation control system thereof |
| CN102307447A (en) * | 2011-08-29 | 2012-01-04 | 中兴通讯股份有限公司 | Heat dissipation method and device |
| CN102520774A (en) * | 2011-11-22 | 2012-06-27 | 浪潮(北京)电子信息产业有限公司 | Radiating system of blade server and radiating control method |
Also Published As
| Publication number | Publication date |
|---|---|
| CN103062091A (en) | 2013-04-24 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN103062091B (en) | A kind of fan intelligent control method | |
| CN101477403B (en) | Automatic control method for system power consumption | |
| CN103345298B (en) | Method of data center energy saving system based on virtual IT resource distribution technology | |
| CN103150003B (en) | The power-saving control method of information handling system, this information handling system and device | |
| CN103645795A (en) | Cloud computing data center energy saving method based on ANN (artificial neural network) | |
| CN106292960A (en) | A kind of server radiating intelligent management | |
| CN104360724B (en) | A kind of heat dissipating method of the blade server based on job scheduling | |
| CN103967823A (en) | Method for managing fan rotating speed of Smart Rack cabinet | |
| CN102520783B (en) | Method capable of realizing energy saving of smart rack and rack system | |
| CN103023802B (en) | A kind of low energy consumption dispatching patcher towards web cluster and method | |
| CN102789301A (en) | Power management method of computer | |
| CN102436297A (en) | Method for saving energy and reducing noise of storage device by use of temperature fusion control | |
| WO2022052479A1 (en) | Power consumption regulation and control method, apparatus and device, and readable storage medium | |
| Yuan et al. | An Online Energy Saving Resource Optimization Methodology for Data Center. | |
| CN103092328A (en) | Computer energy conservation method based on magnetic disc dormancy | |
| CN106484499A (en) | A kind of cloud management platform towards new forms of energy data center extending transversely | |
| CN203673427U (en) | Cooling fan intelligent control system applied to PC-BASED automation equipment | |
| CN203261358U (en) | Genetic algorithm-based cloud computing server cluster | |
| CN116341883B (en) | Resource coordination method and system for photovoltaic grid-connected switch | |
| CN109185208A (en) | A kind of fan speed-governing method and relevant apparatus | |
| CN205038324U (en) | The DC load's automatic loading device | |
| CN103995576A (en) | Computer power supply management method based on CPLD | |
| CN103064504B (en) | Method for energy conservation of main board of server | |
| CN207588480U (en) | The charging pile control system of collective and distributive type control | |
| CN115392684A (en) | Data machine room carbon emission monitoring system and method based on out-of-band mode |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C06 | Publication | ||
| PB01 | Publication | ||
| C10 | Entry into substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| C14 | Grant of patent or utility model | ||
| GR01 | Patent grant | ||
| CF01 | Termination of patent right due to non-payment of annual fee | ||
| CF01 | Termination of patent right due to non-payment of annual fee |
Granted publication date: 20160608 Termination date: 20170128 |