[go: up one dir, main page]

WO2010050080A1 - Ordinateur physique, procédé de commande de dispositif de refroidissement et système serveur - Google Patents

Ordinateur physique, procédé de commande de dispositif de refroidissement et système serveur Download PDF

Info

Publication number
WO2010050080A1
WO2010050080A1 PCT/JP2009/000806 JP2009000806W WO2010050080A1 WO 2010050080 A1 WO2010050080 A1 WO 2010050080A1 JP 2009000806 W JP2009000806 W JP 2009000806W WO 2010050080 A1 WO2010050080 A1 WO 2010050080A1
Authority
WO
WIPO (PCT)
Prior art keywords
temperature
fan
processor
cooling
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.)
Ceased
Application number
PCT/JP2009/000806
Other languages
English (en)
Japanese (ja)
Inventor
陽子 志賀
加藤　猛
高本　良史
林　真一
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
Hitachi Ltd
Original Assignee
Hitachi Ltd
Priority date (The priority date 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 date listed.)
Filing date
Publication date
Application filed by Hitachi Ltd filed Critical Hitachi Ltd
Publication of WO2010050080A1 publication Critical patent/WO2010050080A1/fr
Anticipated expiration legal-status Critical
Ceased legal-status Critical Current

Links

Images

Classifications

    • G—PHYSICS
    • G06—COMPUTING OR CALCULATING; 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 OR CALCULATING; 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
    • G—PHYSICS
    • G06—COMPUTING OR CALCULATING; COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F1/00—Details not covered by groups G06F3/00 - G06F13/00 and G06F21/00
    • G06F1/26—Power supply means, e.g. regulation thereof
    • G06F1/32—Means for saving power
    • G06F1/3203—Power management, i.e. event-based initiation of a power-saving mode
    • 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

Definitions

  • the present invention relates to a control method for a physical computer and a cooling device, and more particularly to a method for controlling an output of a fan and a cooling device of a physical computer in accordance with a CPU operating rate.
  • Patent Document 1 discloses an invention in which the fan is controlled based on the temperature of the heat-generating component and the temperature change to improve the cooling efficiency.
  • Patent Document 2 discloses an invention in which a fan is controlled based on the temperature of a heat-generating component and a temperature change to improve cooling efficiency.
  • Patent Document 3 discloses an invention in which an air flow in a machine room is monitored and ventilation is performed according to the air flow.
  • Patent Document 4 discloses an invention that avoids failure and shutdown by controlling heat dissipation by hibernation when the temperature exceeds a certain value in order to avoid occurrence of a malfunction due to forced shutdown.
  • a control method of the present invention is a control method of a management computer connected to a server device having a processor and a fan and a cooling device, and the temperature and operating rate of the processor from the server
  • the rotation speed of the fan and the intake air temperature to the server are acquired, and a predetermined period has elapsed from the temperature and the operating rate of the processor, the rotation speed of the fan, and the intake air temperature.
  • the estimated temperature of the subsequent processor is calculated, and when the estimated temperature is equal to or higher than a first predetermined value, the target rotational speed of the fan at which the estimated temperature after the period elapses is equal to or lower than the predetermined value.
  • the control method is characterized by determining and instructing the server device to set the target rotational speed.
  • the server system of the present invention has a processor and a fan, and measures the temperature and operating rate of the processor, the rotational speed of the fan, and the intake air temperature, the server device, and the cooling device. And calculating an estimated temperature of the processor after a predetermined period from the temperature and the operating rate of the processor, the rotation speed of the fan, and the intake air temperature, and calculating the estimated A management computer for determining a target rotational speed of the fan at which the estimated temperature is less than or equal to the predetermined value after the period when the temperature is greater than or equal to a first predetermined value.
  • Server system is a processor and a fan, and measures the temperature and operating rate of the processor, the rotational speed of the fan, and the intake air temperature, the server device, and the cooling device. And calculating an estimated temperature of the processor after a predetermined period from the temperature and the operating rate of the processor, the rotation speed of the fan, and the intake air temperature, and calculating the estimated A management computer for determining a target rotational speed of the fan at which the estimated temperature
  • the present invention by cooling the CPU in advance and maintaining the optimum temperature, power consumption due to leakage current can be minimized and the cooling efficiency can be increased.
  • DESCRIPTION OF SYMBOLS 110 Power saving control server, 111 ... Operation information monitoring part, 112 ... Temperature monitoring part, 113 ... CPU temperature estimation part, 114 ... Cooling control determination part, 115 ... Fan rotation speed determination part, 116 ... Fan monitoring / control part, 117: Cooling control unit, 121: Server configuration information, 122: Heat generation profile, 123 ... Fan profile, 124 ... Server operation history, 125 ... CPU temperature profile, 126 ... Rack / cooling map, 127 ... Cooling profile, 200 ... Physical Computer, 223 ... Measurement agent
  • FIG. 1 is a diagram showing a system configuration of an embodiment of the present invention.
  • the system configuration of this embodiment is an information processing system or a storage system.
  • the computer room in which one management computer 100, one or more physical computers 200, one or more storage devices 230, the management computer 100, the physical computers 200, and the storage devices 230 are installed is cooled.
  • a cooling device 151 and a cooling device control unit 150 that controls the cooling device 151 are provided.
  • the management computer 100, the physical computer 200, and the cooling device 151 are connected via the management network 225. Further, the physical computer 200 and the storage device 230 are connected by, for example, a fiber channel network 226.
  • the cooling device control unit 150 may be stored as a program in the memory of the physical computer so as to collectively control the cooling device 151.
  • the cooling device control unit 151 may be stored as a program in a memory in the management computer.
  • FIG. 2 is a diagram showing the management computer 100 in one embodiment of the present invention.
  • the management computer 100 manages the physical computer 200, the storage device 230, and the cooling device control unit 150. Then, information is exchanged with the plurality of physical computers 200 to detect the operation status of the plurality of physical computers 200.
  • the cooling device 151 is individually controlled via the cooling device control function 151 according to the detected operating status of the plurality of physical computers 200. Further, the fan rotation speed and the cooling device of the plurality of physical computers 200 are controlled according to the detected operating status of the plurality of physical computers 200.
  • the management computer 100 includes a central processing unit CPU (Central Processing Unit) 101, a storage device 105 such as a hard disk device or a flash memory, a memory 102, a bus 107, a network interface 104, and a disk interface 103.
  • CPU Central Processing Unit
  • the server 102 is stored in the memory 102.
  • the server program includes an operation information monitoring unit 111, a temperature monitoring unit 112, a CPU temperature estimation unit 113, a fan rotation speed determination unit 115, a fan monitoring / control unit 116, and a cooling control unit 117. These programs are initially stored in the magnetic disk 105, transferred to the memory 102 as necessary, and then executed by the CPU 101.
  • the operation information / power monitoring unit 111 collects operation information and power consumption information of the physical computer 200.
  • the temperature monitoring unit 112 acquires the intake air temperature, CPU temperature, and exhaust temperature of the physical computer.
  • the fan monitoring / control unit 116 acquires information on the fan rotation speed and issues an instruction to change the fan rotation speed.
  • the CPU temperature estimation unit 113 estimates the temperature after a certain time of the CPU built in the physical computer.
  • the fan rotation speed determination unit 115 determines a fan rotation speed that lowers the temperature of the CPU after a predetermined time to a target value.
  • the cooling control determination unit 114 acquires the rack / cooling map 126 and determines the output of the cooling device 151.
  • the cooling control unit 117 issues an instruction to control the output of the cooling device 151.
  • the operation information history 124, the server configuration information 121, the heat generation profile 122, the fan profile 123, the CPU temperature profile 125, the cooling device profile 127, the CPU temperature range 128, and the CPU optimum are stored in the storage device 105.
  • the temperature 129, the rack / cooling map 126, and the cooling device profile 2010 are stored.
  • FIG. 3 is a diagram showing a hardware configuration of the physical computer 200 in one embodiment of the present invention.
  • the physical computer 200 includes a central processing unit CPU 201, a memory 202, a storage device 205 such as a hard disk device or a flash memory, a bus 207, a network interface 204, a disk interface 203, a fan 208, and a BMC (Baseboard Management Controller) 207. .
  • BMC 207 performs monitoring of server inlet temperature, exhaust temperature, CPU temperature, monitoring / control of fan speed, and power supply control.
  • the memory 202 stores an OS 222, a measurement agent program 223 that collects operation information of the physical computer, and a business service program 224. These programs are first stored in the magnetic disk 205, transferred to the memory 202 as necessary, and then executed by the CPU 201. Note that these programs are stored in the magnetic disk 205 by being read from a portable recording medium or downloaded from another computer or storage device via a network connected to each device. It may be a thing.
  • each process of the server program 110 of the management computer is realized by executing each program by a CPU.
  • these are integrated into a processing unit that performs each process, such as a measurement agent determination unit and a measurement unit. Can also be realized in hardware.
  • the measurement agent 223 is a software program that runs on the computer 200 and collects operation information such as a CPU usage rate, a memory usage rate, and a network interface usage rate of a device in which the measurement agent 223 operates and records it as a measurement counter.
  • the operation information / power monitoring unit 111 of the server program 110 of the management computer transmits an operation information collection request by SNMP (Simple Network Management Protocol) to the measurement agent 223.
  • the measurement agent 223 receives this operation information collection request, and transmits the value of the measurement counter designated by the object ID (Identification) in the request to the operation information / power monitoring unit 111.
  • the server program 110 can centrally manage the operation information of a plurality of management targets by receiving the value of the measurement counter and recording it as operation information.
  • FIG. 4 is a diagram showing a device arrangement of the computer room 400 in which the physical computer 200, the storage device 230, the cooling device 151, and the like are installed according to an embodiment of the present invention.
  • each rack 401 and the cooling devices 151a and 151b are fixed on the floor.
  • a plurality of outlets 431 to 435 are installed on the floor.
  • the motor 440 is fixed to the air outlets 431 to 435, and an opening / closing plate 442 that opens and closes each air outlet according to the rotational drive of the motor 440 is provided on the motor rotating shaft 441. It is a fixed configuration.
  • the physical computer 200 and the storage device 230 are stored in the rack 401a.
  • the racks 401b to 401c similarly store the physical computer 200 and the storage device 230 (not shown).
  • the cooling devices 151 a and 151 b are attached to the side surface of the computer room 400 and are configured as one element of the cooling device 151 for keeping the temperature of the computer room 400 constant.
  • the cooling devices 151 a and 151 b remove the heat discharged from each server by sending cold air under the floor and blowing out the cold air from the air outlet (perforated tile) 431. At this time, in accordance with an instruction from the management computer 100, control is performed to open one of the outlets 431 to 435 and close the other outlets.
  • the blower outlet 431 is driven by the rotation of the motor 440 as a control for the cooling device 151.
  • the air outlet 433 is closed and the other air outlets 431, 432, 434, 435 are opened.
  • the air outlets 431 to 431 are controlled by the rotational drive of the motor 440 as a control for the cooling device 151. Control is performed such that the air outlets 431 and 435 out of 435 are closed and the other air outlets 432, 433, and 434 are opened.
  • the cooling device in the present embodiment is a general computer room air conditioner (CRAC: “Computer” Room ”Air” Conditioner), but is not limited thereto.
  • the cooling facility may be a liquid cooling device that removes heat discharged from each server by circulating the cooled liquid refrigerant through the pipes and circulating through each rack. In the liquid cooling device, there is a valve in front of the pipe that leads to each rack, and the cooling output is adjusted by opening and closing the valve in the same manner as the outlet. Further, the cooling facility may be an outside air cooling device that removes heat discharged from each server by taking in outside cold air and sending cool air from under the floor in the same manner as the computer room cooling device.
  • FIG. 5 is a diagram showing the server configuration information 121 in one embodiment of the present invention.
  • the server configuration information 121 includes a rack / physical computer map 500 (FIG. 5 (a)) and a physical computer list 510 (FIG. 5) showing the correspondence between racks installed in the computer room 400 and physical computers stored in the racks. 5 (b)).
  • the rack / physical computer map 500 (FIG. 5A) includes a rack ID 501 that is an identifier of the rack and a physical computer ID 502 that is an identifier of the physical computer 200 stored in each rack.
  • the physical computer list 510 (FIG. 5B) is one or more records including a physical computer ID 511, a chassis number 512, a component identifier (item) 513, and a component value 514 that the physical computer 200 has. Configured and represents the processing capability of the physical computer 200.
  • the physical computer ID 501 stores an identifier of each physical computer.
  • the chassis number 512 is used to specify the chassis that stores the blade server when the physical computer is a blade server. In the case of a non-modular type server such as a 1U server, “-” is recorded.
  • a blade server shares a fan and a power source among a plurality of servers, and may have a management processor that manages server configuration and power on / off.
  • the management target physical computer is a blade server
  • the IP address and the port number necessary for connecting to the management processor are managed as server configuration information (not shown).
  • the server configuration information 121 is often determined by a designer of a managed system at the time of system construction and managed by a document or software.
  • the physical computer configuration information may be created based on such managed configuration information, or may be created from dynamically collected information.
  • FIG. 6 is a diagram showing a rack / cooling device map 126 according to an embodiment of the present invention.
  • the rack / cooling device map 126 includes one or more records including a rack identifier 601, an identifier 602 of the cooling device 151, an outlet 603 located on the front of the rack, and an outlet 604 located on the rear of the rack. Consists of. Each record represents a correspondence between a rack, a cooling device 151 that cools each rack, and an identification number of an outlet that blows and ventilates the rack.
  • FIG. 7 is a diagram showing operation information 710 (FIG. 7A) and power information 720 (FIG. 7B) in an embodiment of the present invention.
  • the operation information 710 indicates the resource usage status of one physical computer 200. As an example, it is composed of one or more records including a measurement date 711, a measurement day 712, a measurement time 713, and a CPU operation rate 714. The unit of the CPU operation rate is%.
  • the operation information shown here can be acquired by WMI (Windows Management Interface) in the case of Windows (registered trademark) and by the Top command in the case of Linux.
  • the power information 720 (FIG. 7B) indicates the power consumption status of each physical computer 200.
  • the measurement date 721, the measurement day 722, the measurement time 723, the physical computer power amount 724, and the chassis power amount. 726 is composed of one or more records.
  • the physical computer 200 is a blade server, the power amount of a plurality of physical computers and the power amount of the chassis are managed in one table.
  • the physical computer 200 is not a blade server, only the power amount of one physical computer is managed.
  • FIG. 8 is a diagram showing a heat generation profile 122 in one embodiment of the present invention.
  • the heat generation profile 122 is composed of one or more records including a CPU operation rate 801 and a heat generation amount 802.
  • the operation rate 811 is the operation rate of the CPU
  • the heat generation amount and 802 are the heat generation amount of the CPU. That is, each record represents the amount of heat generated with respect to the operating rate of the CPU of the physical computer 200.
  • the heat generation profile 122 is different for each type of CPU included in the physical computer 200. There are various methods for obtaining the heat generation profile. For example, it is possible to record a past operating rate and a calorific value history of the CPU, and obtain from the history. Moreover, the process by CPU can be performed in advance and the relationship between an operation rate and the emitted-heat amount can also be measured. It is also possible to record the relationship between the operating rate provided by the CPU vendor and the heat generation amount.
  • FIG. 9 shows the CPU temperature profile 125 (FIG. 9A), the CPU temperature range 128 (FIG. 9B), the CPU optimum temperature 129 (FIG. 9C), the temperature rise in one embodiment of the present invention. It is a figure which shows the relationship of the power consumption by leak current, and the relationship (FIG.9 (d)) of temperature rise and fan power consumption.
  • the CPU temperature profile 125 (FIG. 9A) is composed of one or more records including a CPU heat generation amount 901 and a CPU temperature change 902 per fixed time. That is, each record represents a temperature change amount after a certain time with respect to the heat generation amount of the CPU.
  • the temperature change of an object is a value obtained by dividing the amount of heat given from the outside by the heat capacity, but the ease of heat transfer differs depending on the nature of the object. That is, even when the same amount of heat is given, the temperature finally reached and the rate of temperature change differ depending on the material and structure of the CPU. Therefore, the CPU temperature profile is a table determined for each CPU type.
  • the CPU temperature profile 125 is different for each type of CPU included in the physical computer 200.
  • the CPU temperature range 128 (FIG. 9B) includes an upper limit value 911 and a lower limit value 912, and indicates a temperature range where the CPU operates safely.
  • the upper and lower limits of the temperature at which the CPU operates normally are determined by the manufacturing vendor. In many servers, when the CPU temperature exceeds this range due to a rise in room temperature or a fan failure, a program for monitoring the hardware state of the server issues a warning to the outside of the server.
  • the CPU optimum temperature 129 is a temperature that minimizes the sum of the power consumed by the leakage current of the CPU and the power consumed by the fan that cools the CPU. Since the CPU optimum temperature 129 varies depending on the intake air temperature, the CPU optimum temperature 129 is composed of one or more records including the intake air temperature 921 and the CPU optimum temperature 922.
  • the CPU optimum temperature 129 will be described.
  • a small amount of current (leakage current) flows through the CPU even when it is in an OFF state due to miniaturization of the semiconductor, but this leakage current has a characteristic that it rises exponentially as the temperature rises. For this reason, even in the idle state, when the temperature of the CPU increases, the power consumption also increases exponentially.
  • fan power consumption is proportional to the fan wind speed and power consumption, and inversely proportional to the square of the CPU temperature rise. Therefore, as shown in FIG.
  • the leakage current can be suppressed if the temperature is kept low, but the power consumption of the fan increases. If the temperature rise is allowed, the leakage current increases, but the power consumption of the fan is low. It is a trade-off relationship that can be done.
  • the CPU optimum temperature 129 is a temperature at which the sum of the power consumed by the leak current and the power consumed by the fan is minimized, can be measured by the administrator, and can be disclosed by the server vendor.
  • FIG. 10 is a diagram showing a fan profile 123 in one embodiment of the present invention.
  • the fan profile 123 is composed of one or more records each including a fan rotation speed 1001 of the physical computer 200 and a CPU temperature change 1002 that can be changed per unit time by cooling the CPU by blowing air from the fan. .
  • the cooling efficiency of CPU by a fan changes with inlet temperature. Therefore, the temperature change of the CPU that can be changed per certain time by the fan differs depending on the intake air temperature.
  • the fan profile of the present embodiment shows the case of the inlet air temperature of 21 ° C., 22 ° C., and 23 ° C. (1003), it is not limited to this.
  • the relationship between the server CPU and the fans is not necessarily one-to-one.
  • the plurality of servers may be cooled by a shared fan.
  • the fan is configured to cool each CPU uniformly, and the fan profile 123 as described above can be defined.
  • FIG. 20 is a diagram showing a cooling device profile 2010 and a cooling control pattern 2010 in one embodiment of the present invention.
  • the cooling control profile 2010 is divided for each cooling device, and includes one or more records including a cooling device output stage 2011, an inlet air temperature change 2012 of each cooling target rack at the output stage, and power consumption 2013. Consists of.
  • the cooling control pattern 2020 represents the combination of the outputs of the plurality of cooling devices and the power consumption, and includes a combination number 2021, the output 2022 of each cooling device, and the power consumption 2023 of the entire cooling device at that time. Consists of the above records.
  • FIGS. 5 to 10 The information described in FIGS. 5 to 10 is described in the definition file by the administrator. However, these pieces of information may be input from a GUI (Graphical User Interface) instead of a definition file, or may be acquired from another server via a network.
  • GUI Graphic User Interface
  • FIG. 11 is a diagram showing a control flow of the physical computer and the cooling device by the management computer in one embodiment of the present invention.
  • the operation information monitoring unit 111 of the server program 110 of the physical computer reads the server configuration information 121 from the storage device 105 (S1101). Then, the physical computer 200 to be managed is grasped, and the operation information monitoring unit 111 collects operation information and power consumption of the physical computer 200 (S1102) and stores them in the server operation history 124.
  • the CPU temperature estimation unit 113 estimates the CPU temperature of each physical computer 200 after a certain period of time based on the stored operation history, the heat generation profile 122 and the CPU temperature profile 125 stored in the storage device ( S1103). This estimation process will be described with reference to FIG.
  • FIG. 12 is a diagram showing a CPU temperature estimation flow in one embodiment of the present invention. This process is executed by the CPU temperature estimation unit.
  • the CPU temperature estimation unit 113 refers to the server operation history 124 and acquires the CPU operation rate 714 of the physical computer 200 that is the processing target (S1201).
  • the amount of heat generated with respect to the CPU operating rate is obtained with reference to the heat generation profile 122 (S1202).
  • the CPU type of the physical computer 200 is obtained by referring to the server list 510 of the server configuration information 121, and the CPU temperature profile corresponding to the CPU type is used.
  • the CPU temperature change with respect to the generated heat quantity is obtained with reference to the CPU temperature profile 125 (S1203).
  • the CPU temperature estimation unit 113 acquires the current CPU temperature and the inlet temperature to the server device using the temperature monitoring unit, and uses the fan monitoring / control unit 116 to determine the current fan rotation speed. get.
  • the CPU temperature change (cooling effect) in the case where the current fan speed is maintained from the current time until a certain time has elapsed is obtained (S1204).
  • the fan monitoring / control unit 116 is connected to the BMC 207 by SSH (Secure Shell) and executes a command for acquiring the fan rotation speed, and acquires the value of the fan rotation speed.
  • SSH Secure Shell
  • the temperature monitoring unit acquires the intake air temperature to the physical computer, and the fan rotation speed determination unit 115 refers to the acquired intake air temperature and the fan profile 123. Then, the number of fan rotations necessary to keep the CPU temperature within the upper limit value within a certain time is calculated (S1105).
  • FIG. 13 is a diagram showing a fan rotational speed determination flow in one embodiment of the present invention.
  • the fan rotational speed determination unit 115 obtains the difference between the estimated CPU temperature and the threshold value of the CPU temperature after the elapse of a predetermined time, and the CPU temperature change amount to be realized. (S1301).
  • the process waits for a certain period of time (S1108) and returns to monitoring of operation information, temperature, and rotation speed (S1102).
  • the threshold value may be an upper limit value of the CPU temperature range 128.
  • the threshold value may be a value obtained by subtracting a certain value from the upper limit value. By setting a value obtained by subtracting a constant value from the upper limit value as a threshold value, the upper limit value can be operated even when the operation amount suddenly increases.
  • whether or not to control the rotation speed of the fan is determined depending on whether or not the threshold value after a certain time has elapsed, but the present invention is not limited to this.
  • the inlet temperature is acquired by the temperature monitoring unit 112, and the CPU optimum temperature 129 that is a value that minimizes the sum of the CPU leakage current and the fan power consumption is determined from the acquired inlet temperature. Then, the determination 1104 may be made based on whether the value obtained by subtracting the constant temperature from the CPU optimum temperature is within the range of the value obtained by adding the constant temperature to the CPU optimum temperature.
  • the server program 110 confirms whether or not the physical computer 200 can change the rotational speed of the fan to the calculated rotational speed (S1106). For example, it is confirmed whether the fan rotation speed obtained in the process 1105 exceeds the maximum value. If it can be changed (S1106: Y), the fan monitoring / control unit 116 instructs the physical computer 200 to change the rotational speed of the fan (S1107). Then, after a predetermined time has elapsed (S1108), the process returns to monitoring of operation information, temperature, and rotation speed (S1102).
  • the cooling control unit 117 sets the cooling facilities 151a and 151b installed in the computer room 400. And the inlet temperature of the physical computer 200 is lowered (S1520). This process will be described with reference to FIG.
  • FIG. 21 is a diagram for explaining a cooling control flow in one embodiment of the present invention.
  • the cooling control unit 117 refers to the fan profile 123 to obtain a target inlet temperature that achieves the CPU temperature change amount to be realized (S2101).
  • the fan rotation speed is the maximum value that can be realized. That is, if the 5000 revolutions in FIG. 10 is the maximum value and the CPU temperature change is ⁇ 3.0 ° C., the target inlet air temperature is obtained as 21 ° C.
  • the cooling control unit 117 refers to the server configuration information 121 to identify the rack in which the physical computer 200 to be controlled is stored. Then, referring to the rack / cooling map 126, the cooling facility responsible for cooling the specified rack is specified. Multiple racks may be specified. And the cooling control part 117 instruct
  • the specific method for determining the output of the cooling device refers to the cooling device profile 2010 and selects an output stage in which the rack temperature change amount is equal to or greater than the intake air temperature change target value.
  • a cooling control pattern list 2030 listing the combinations of outputs that the cooling equipment 151a and the cooling equipment 151b can take is created (S2103).
  • the cooling equipment 151a is set to stage 1, the cooling equipment 151a is set to stage 3, the both outputs are set to stage 2, the cooling equipment 151a is set to stage 3, and the cooling equipment 151b is set to stage 1.
  • There are patterns such as.
  • the power consumption of the cooling device with respect to the temperature change that is, the cooling temperature varies depending on the characteristics of the device and the distance to the rack, and therefore the power consumption of each pattern is different.
  • the output of the cooling equipment 151a, 151b is changed to lower the inlet temperature of the physical computer 200 (S2105).
  • the operation information / temperature / revolution speed monitoring is resumed (S1102).
  • the physical computer profile 124 of the present embodiment focuses only on the CPU that is the main heat generating component, but the heat generation profile may be information corresponding to the utilization rate of other components in addition to the CPU.
  • the IT device may be a storage device or a network device.
  • the IT device is a storage device
  • the amount of heat generated by the device changes not only with the CPU operation rate but also with IOPS (InputInOutput Per Second) indicating the number of data input / output to the device.
  • IOPS InputInOutput Per Second
  • the calorific value can be estimated based on the above.
  • the heat generation amount can be estimated based on the port usage rate.
  • the power saving control server 110 includes the operation information monitoring unit 111 that collects the operation information and power consumption information of the physical computer 200, and the temperature that collects the inlet temperature, CPU temperature, and exhaust temperature of the managed server.
  • the monitoring unit 112 monitors the fan rotation speed of the management target server, reads the fan monitoring / control unit 116 that changes the rotation speed of the fan, the heat generation profile 122, and the CPU temperature profile 125, and the management target server
  • a CPU temperature estimation unit 113 that estimates the temperature after a certain time of the built-in CPU
  • a fan rotation number determination unit 115 that reads the fan profile 123 and determines a fan rotation number that lowers the temperature after a certain time of the CPU to a target value.
  • Cooling device profile 127 and rack / cooling map 126 are received and cooling
  • the cooling control determining unit 114 for determining the output of the device 151 and the cooling control unit 117 for instructing the control of the cooling device 151 are used to cool the CPU in advance to maintain the optimum temperature, and to reduce the power generated by the leakage current. Consumption can be minimized and cooling efficiency can be increased.
  • the server fan and air conditioning in cooperation, even if the CPU temperature cannot be within the specified range only by controlling the fan, the air intake temperature is lowered in advance by adjusting the air conditioning. As a result, the temperature of the CPU is kept within a specified range, and the occurrence of a failure due to heat can be avoided.
  • the BMC in the server device controls the fan according to an instruction from the management computer.
  • the server program 110 is stored in the memory of the physical computer, and the operation information history is recorded. 124, server configuration information 121, heat generation profile 122, fan profile 123, CPU temperature profile 125, cooling device profile 127, CPU temperature range 128, CPU optimum temperature 129, rack / cooling map 126
  • the cooling device profile 2010 may be stored in the storage device, and the fan control and the air conditioning device control in this embodiment may be performed in the physical computer.
  • the system configuration of one embodiment of the present invention is the same as in FIG.
  • the physical computer is the same as that shown in FIG.
  • FIG. 14 is a diagram showing a management computer according to an embodiment of the present invention.
  • the difference from the management computer of the first embodiment is that the operation history information 124 is not held, but the job execution schedule 132 and the server / job map 131 are held.
  • Description of the server configuration information 121, the heat generation profile 122, the fan profile 123, the CPU temperature profile 125, the rack / cooling map 126, the CPU temperature range 128, and the CPU optimum temperature 129 already described in the first embodiment will be omitted.
  • the server program 110 obtains the CPU operating rate in a certain period from the job execution schedule (load fluctuation) of the business executed by the managed server, and based on the CPU operating rate and the heat generation amount for the operating rate determined for each CPU type.
  • the temperature of the CPU after a certain time is estimated and it is confirmed that the CPU temperature exceeds the upper limit value of the temperature range in which the CPU operates stably, the fan speed of the managed server and the output of the cooling device as necessary To control.
  • FIG. 15 is a configuration diagram showing a server / job map 131 (FIG. 15A) and a job execution schedule 132 (FIG. 15B) according to an embodiment of the present invention.
  • the server job map 131 (FIG. 15A) is composed of one or more records including a physical computer ID 1401 that is an identifier of the physical computer and a business type 1402 that indicates the type of business. Each record indicates which business is being executed on which physical computer.
  • the job execution schedule 132 (FIG. 15B) is one or more records including a day 1411, a day of the week 1412, a start time 1413, an end time 1414, a job ID 1415, and an average CPU operation rate 1416 for each physical computer. Composed. Each record indicates the average value of the CPU operation rate generated by the business for each day of the week and time zone.
  • the CPU operation rate of the server is predicted based on the batch job execution schedule, the time variation of the business request, or the server On / Off schedule.
  • the job execution schedule 132 is obtained from the server operation history when processing equivalent to the batch job execution schedule, the time variation of the business request, or the server On / Off schedule is executed.
  • FIG. 16 is a flowchart showing the control flow of the second embodiment of the present invention.
  • the operation information monitoring unit 111 of the server program 110 acquires server configuration information 121 (S1601), and acquires information on the physical computer 200 to be managed. Then, the job execution schedule information of the physical computer to be managed is acquired.
  • the server program 110 confirms the job execution schedule and confirms whether the end time of the job executed by the physical computer 200 has been reached (S1603). If it is the job end time (S1603: Y), the server / job map 131 is referenced to identify the next job to be executed by the physical computer 200, and the job execution schedule 132 corresponding to the job is referred to. The start and end times of jobs executed by the physical computer 200 and the average CPU operating rate are acquired (S1604).
  • the CPU temperature estimation unit 113 estimates the CPU temperature of each physical computer 200 after a certain period of time based on the average CPU operating rate, the heat generation profile 122, and the CPU temperature profile 125 (S1604).
  • the temperature estimation process is the same as in the first embodiment.
  • the power consumption can be suppressed by reducing or stopping the rotation speed of the fan.
  • processing of processing 1606 is the same as that of the first embodiment.
  • the job to be executed and the CPU operating rate for processing the job are acquired from the job execution schedule.
  • the job execution schedule it is not limited to this, for example, storing data such as jobs processed in the past and CPU utilization rate, predicting CPU utilization rate for a certain time based on the stored data, CPU utilization may be acquired.
  • the system configuration of one embodiment of the present invention is the same as in FIG.
  • the physical computer is the same as that shown in FIG.
  • FIG. 17 is a diagram showing a management computer according to an embodiment of the present invention.
  • the difference from the management computer of the first embodiment is that the server program holds the rule determination unit 133 and the storage device holds the cooling control rule.
  • the server configuration information 121, heat generation profile 122, fan profile 123, operation history information 124, CPU temperature profile 125, rack / cooling map 126, CPU temperature range 128, and CPU optimum temperature 129 already described in the first embodiment Description is omitted.
  • FIG. 18 is a configuration diagram showing a cooling control rule 131 of the third embodiment of the present invention.
  • the cooling control rule 131 includes a rule 1010 and an action 1020.
  • the rule 1010 includes one or more records including an evaluation item 1011 and a threshold value 1012. Each record represents one condition. The condition is satisfied when the value of the evaluation item is equal to or greater than the threshold value, and is specified by the action 1020 when the condition expressed by all the records is satisfied. Control is executed. Further, when the condition of one item is satisfied, the control designated by the action 1020 may be executed.
  • the cooling control rule is a condition that can be determined that an abnormal temperature rise has started, and is defined in advance by a computer room administrator.
  • the CPU temperature of all the physical computers 200 stored in a certain rack exceeds 60 ° C.
  • the exhaust temperature exceeds 40 ° C.
  • the fan rotation speed Is more than 10,000 revolutions / second, it is considered that a heat pool is generated on the rear surface of the rack where the exhaust of the physical computer 200 stored in the rack comes out.
  • action 1020 is executed to discharge this heat. It is instructed to maximize the output of the cooling device 151 for cooling the rack and to open 100% of the grating plate at the outlet located on the back of the rack.
  • the intake air temperature of all the physical computers 200 stored in the same rack exceeds a certain threshold value, it is considered that a heat accumulation has occurred on the front surface of the rack, and the cooling device 151 for cooling the rack. It is also effective to open the grating plate at the outlet located in front of the rack 100%.
  • FIG. 19 is a flowchart showing a control flow of the third embodiment of the present invention.
  • the operation information monitoring unit 111 of the power saving control server 110 acquires the server configuration information 121 (S1901).
  • the operation information monitoring unit 111 collects operation information and power consumption of the physical computer 200 and stores them in the server operation history 124.
  • the temperature monitoring unit 112 collects the current CPU temperature and exhaust temperature.
  • the fan monitoring / control unit 116 collects the fan rotation speed (S1902).
  • the rule determination unit 133 compares the value of each item of the rule 1010 of the cooling control rule 131 with the threshold value 1012 for the physical computer 200 stored in each rack based on the collected information (S1903). ).
  • the rack / cooling map 126 is referred to, the cooling device 151 that cools the rack, and the The air outlet located on the back of the rack is specified, and the cooling control designated by action 1020 is executed (S1905).
  • the process returns to monitoring of operation information and the like (S1902).
  • the present embodiment it is possible to reduce the installation cost of the sensor as compared with a method of detecting a temperature increase by installing a large number of sensors around the management target device.

Landscapes

  • Engineering & Computer Science (AREA)
  • Theoretical Computer Science (AREA)
  • Physics & Mathematics (AREA)
  • General Engineering & Computer Science (AREA)
  • General Physics & Mathematics (AREA)
  • Human Computer Interaction (AREA)
  • Computer Hardware Design (AREA)
  • Thermal Sciences (AREA)
  • Microelectronics & Electronic Packaging (AREA)
  • Power Sources (AREA)
  • Cooling Or The Like Of Electrical Apparatus (AREA)

Abstract

L'invention porte sur la détermination d'une commande de refroidissement pour réduire la consommation d'énergie globale d'un système. L'invention porte sur un procédé pour commander un ordinateur de gestion connecté à un dispositif serveur comportant un processeur et un ventilateur et un dispositif de refroidissement caractérisé en ce que la température et la fréquence de fonctionnement du processeur, la vitesse de rotation du ventilateur et la température de l'air envoyé dans le serveur sont acquis à partir du serveur, la température du processeur après un laps de temps prédéterminé est estimée par calcul sur la base de la température et de la fréquence de fonctionnement du processeur, de la vitesse de rotation du ventilateur et de la température de l'air envoyé, la vitesse de rotation cible du ventilateur à laquelle la température estimée après le laps de temps est inférieure à une première valeur prédéterminée est déterminée si la température estimée est supérieure à la première valeur déterminée, et il est ordonné au dispositif serveur de changer la vitesse de rotation pour la vitesse de rotation cible.
PCT/JP2009/000806 2008-10-31 2009-02-24 Ordinateur physique, procédé de commande de dispositif de refroidissement et système serveur Ceased WO2010050080A1 (fr)

Applications Claiming Priority (2)

Application Number Priority Date Filing Date Title
JP2008-280698 2008-10-31
JP2008280698A JP2010108324A (ja) 2008-10-31 2008-10-31 物理計算機及び冷却装置の制御方法及びサーバシステム

Publications (1)

Publication Number Publication Date
WO2010050080A1 true WO2010050080A1 (fr) 2010-05-06

Family

ID=42128456

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/JP2009/000806 Ceased WO2010050080A1 (fr) 2008-10-31 2009-02-24 Ordinateur physique, procédé de commande de dispositif de refroidissement et système serveur

Country Status (2)

Country Link
JP (1) JP2010108324A (fr)
WO (1) WO2010050080A1 (fr)

Cited By (10)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
JP2011196617A (ja) * 2010-03-19 2011-10-06 Fujitsu Ltd 空調システム及び空調方法
CN102314206A (zh) * 2010-07-06 2012-01-11 英业达股份有限公司 服务器的风扇控速装置
CN102478935A (zh) * 2010-11-29 2012-05-30 英业达股份有限公司 机架式服务器系统
CN103883851A (zh) * 2014-04-18 2014-06-25 沈潇 一种风冷摆动的笔记本电脑支座
US9341190B2 (en) 2012-10-18 2016-05-17 International Business Machines Corporation Thermal control system based on nonlinear zonal fan operation and optimized fan power
US9354126B2 (en) 2012-11-30 2016-05-31 International Business Machines Corporation Calibrating thermal behavior of electronics
US10180665B2 (en) 2011-09-16 2019-01-15 Lenovo Enterprise Solutions (Singapore) Pte. Ltd. Fluid-cooled computer system with proactive cooling control using power consumption trend analysis
CN115467849A (zh) * 2022-10-14 2022-12-13 济南浪潮数据技术有限公司 一种风扇控制方法及相关装置
WO2023078237A1 (fr) * 2021-11-02 2023-05-11 北京百度网讯科技有限公司 Procédé et appareil pour réguler la température d'un serveur de téléphone mobile en nuage, et dispositif
US12513851B2 (en) 2021-10-21 2025-12-30 Canon Kabushiki Kaisha Electronic device and control method

Families Citing this family (25)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
JP5197675B2 (ja) 2010-05-14 2013-05-15 株式会社東芝 空調システム
JP2012053678A (ja) * 2010-09-01 2012-03-15 Fujitsu Ltd ファン制御プログラム、ファン制御方法及び情報処理装置
JP5545137B2 (ja) * 2010-09-03 2014-07-09 富士通株式会社 電子機器、制御プログラムおよびシャットダウン制御方法
CN102478937A (zh) * 2010-11-30 2012-05-30 英业达股份有限公司 机架服务器系统
JP5511698B2 (ja) * 2011-01-20 2014-06-04 日本電信電話株式会社 空調機連係制御システム、空調機連係制御方法および空調機連係制御プログラム
JP5672099B2 (ja) * 2011-03-22 2015-02-18 富士通株式会社 電子機器を搭載した装置、電子機器を搭載した装置の冷却プログラム及び電子機器を搭載した装置の冷却方法
JP5691933B2 (ja) 2011-08-16 2015-04-01 富士通株式会社 空調制御方法、空調制御システム及び空調制御装置
WO2013038470A1 (fr) 2011-09-12 2013-03-21 富士通株式会社 Système de refroidissement, procédé de refroidissement et programme de commande de refroidissement
JP5568535B2 (ja) * 2011-09-28 2014-08-06 株式会社日立製作所 データセンタの負荷割当て方法及び情報処理システム
WO2013101116A1 (fr) * 2011-12-29 2013-07-04 Intel Corporation Étranglement thermique adaptatif avec capacité de configuration utilisateur
EP2833238A4 (fr) * 2012-03-30 2015-03-11 Fujitsu Ltd Dispositif de traitement d'informations, procédé de commande et programme
JP2013213635A (ja) * 2012-04-03 2013-10-17 Nippon Telegr & Teleph Corp <Ntt> 空調制御方法および空調制御システム
JP6036428B2 (ja) * 2013-03-18 2016-11-30 富士通株式会社 電子機器冷却システム及び電子機器冷却方法
JP6020283B2 (ja) * 2013-03-26 2016-11-02 富士通株式会社 電子機器冷却システム
JP6083305B2 (ja) * 2013-04-08 2017-02-22 富士通株式会社 電子機器冷却システム
JP6002098B2 (ja) * 2013-07-23 2016-10-05 日本電信電話株式会社 空調制御方法および空調制御システム
JP6417672B2 (ja) 2014-02-27 2018-11-07 富士通株式会社 データセンタ、データセンタの制御方法及び制御プログラム
JP6287434B2 (ja) * 2014-03-26 2018-03-07 日本電気株式会社 温度制御装置、温度制御方法、及び温度制御プログラム
JP6384321B2 (ja) 2014-12-26 2018-09-05 富士通株式会社 ジョブ割当プログラム、方法及び装置
JP6628311B2 (ja) 2016-03-24 2020-01-08 Ｎｅｃプラットフォームズ株式会社 ファン制御装置、冷却ファン・システム、コンピュータ装置、ファン制御方法及びプログラム
JP7037057B2 (ja) * 2018-06-29 2022-03-16 富士通株式会社 電子装置、制御プログラムおよび制御方法
JP7213049B2 (ja) * 2018-09-28 2023-01-26 中央電子株式会社 ファン制御装置、制御方法、制御プログラム、サーバーラック、および空調管理システム
JP7292908B2 (ja) * 2019-03-13 2023-06-19 株式会社東芝 異常検出装置、異常検出方法、およびプログラム
JP6659896B2 (ja) * 2019-05-20 2020-03-04 株式会社Ｎｔｔファシリティーズ 空調システム
JP2023014532A (ja) * 2021-07-19 2023-01-31 オムロン株式会社 画像処理装置および制御方法

Citations (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
JP2002175131A (ja) * 2000-12-07 2002-06-21 Nec Yonezawa Ltd 情報処理装置
JP2006208000A (ja) * 2005-01-28 2006-08-10 Hewlett-Packard Development Co Lp 熱／電力管理装置
JP2007041739A (ja) * 2005-08-02 2007-02-15 Nec Access Technica Ltd ファン駆動制御方法及び装置
US20070067136A1 (en) * 2005-08-25 2007-03-22 Conroy David G Methods and apparatuses for dynamic thermal control

Patent Citations (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
JP2002175131A (ja) * 2000-12-07 2002-06-21 Nec Yonezawa Ltd 情報処理装置
JP2006208000A (ja) * 2005-01-28 2006-08-10 Hewlett-Packard Development Co Lp 熱／電力管理装置
JP2007041739A (ja) * 2005-08-02 2007-02-15 Nec Access Technica Ltd ファン駆動制御方法及び装置
US20070067136A1 (en) * 2005-08-25 2007-03-22 Conroy David G Methods and apparatuses for dynamic thermal control

Cited By (13)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
JP2011196617A (ja) * 2010-03-19 2011-10-06 Fujitsu Ltd 空調システム及び空調方法
CN102314206A (zh) * 2010-07-06 2012-01-11 英业达股份有限公司 服务器的风扇控速装置
CN102478935A (zh) * 2010-11-29 2012-05-30 英业达股份有限公司 机架式服务器系统
US10180665B2 (en) 2011-09-16 2019-01-15 Lenovo Enterprise Solutions (Singapore) Pte. Ltd. Fluid-cooled computer system with proactive cooling control using power consumption trend analysis
US9360021B2 (en) 2012-10-18 2016-06-07 International Business Machines Corporation Thermal control system based on nonlinear zonal fan operation and optimized fan power
US9341190B2 (en) 2012-10-18 2016-05-17 International Business Machines Corporation Thermal control system based on nonlinear zonal fan operation and optimized fan power
US9354126B2 (en) 2012-11-30 2016-05-31 International Business Machines Corporation Calibrating thermal behavior of electronics
US9534967B2 (en) 2012-11-30 2017-01-03 International Business Machines Corporation Calibrating thermal behavior of electronics
US9702767B2 (en) 2012-11-30 2017-07-11 International Business Machines Corporation Calibrating thermal behavior of electronics
CN103883851A (zh) * 2014-04-18 2014-06-25 沈潇 一种风冷摆动的笔记本电脑支座
US12513851B2 (en) 2021-10-21 2025-12-30 Canon Kabushiki Kaisha Electronic device and control method
WO2023078237A1 (fr) * 2021-11-02 2023-05-11 北京百度网讯科技有限公司 Procédé et appareil pour réguler la température d'un serveur de téléphone mobile en nuage, et dispositif
CN115467849A (zh) * 2022-10-14 2022-12-13 济南浪潮数据技术有限公司 一种风扇控制方法及相关装置

Also Published As

Publication number Publication date
JP2010108324A (ja) 2010-05-13

Similar Documents

Publication Publication Date Title
WO2010050080A1 (fr) Ordinateur physique, procédé de commande de dispositif de refroidissement et système serveur
JP4922255B2 (ja) 情報処理システムおよびそのシステムにおける省電力制御方法
US8341433B2 (en) Method and system for managing the power consumption of an information handling system
US7991515B2 (en) Computer cooling system with preferential cooling device selection
US8909384B1 (en) Computing apparatus operable under multiple operational policies
US10817039B2 (en) Adjusting a power limit in response to a temperature difference
US9671839B2 (en) Information handling system dynamic acoustical management
US8676397B2 (en) Regulating the temperature of a datacenter
CN101032196B (zh) 使用传热水平操作的能量高效crac部件及方法
JP5835465B2 (ja) 情報処理装置、制御方法、及びプログラム
US9723763B2 (en) Computing device, method, and computer program for controlling cooling fluid flow into a computer housing
JP2008235696A (ja) ファン回転制御方法、ファン回転制御システム、およびファン回転制御プログラム
CN102197385A (zh) 根据环境条件执行基于区域的工作负荷调度
JP5921461B2 (ja) 外気冷却および局所冷却式情報処理システムとその負荷割当て方法
KR20140092328A (ko) 누설 전류 측정으로부터 열 관리 폴리시를 결정하는 시스템 및 방법
JP5969939B2 (ja) データセンタの空調制御装置
EP2575003B1 (fr) Procédé pour déterminer l&#39;attribution de charges de centre de données et système de traitement d&#39;informations
US8630739B2 (en) Exergy based evaluation of an infrastructure
CN107300278B (zh) 用于最小化hvac系统中的压缩机使用的系统和方法
JP5165104B2 (ja) 情報処理システムおよびそのシステムにおける省電力制御方法
Chaudhry et al. Considering thermal-aware proactive and reactive scheduling and cooling for green data-centers
US20210271300A1 (en) Dynamic thermal control
JP2017084013A (ja) 電力管理プログラム、電力管理方法、及び電力管理装置

Legal Events

Date Code Title Description
121 Ep: the epo has been informed by wipo that ep was designated in this application

Ref document number: 09823197

Country of ref document: EP

Kind code of ref document: A1

NENP Non-entry into the national phase

Ref country code: DE

122 Ep: pct application non-entry in european phase

Ref document number: 09823197

Country of ref document: EP

Kind code of ref document: A1