Cooling fan control device, cooling fan control method, electronic apparatus, and program

Abstract

A cooling fan control device (10) controls a cooling fan (23). The cooling fan control device (10) comprises: a power consumption estimation unit (12) that, on the basis of measurement information pertaining to power consumption in a unit (21) to be cooled that is measured by a sensor unit (22), estimates the power consumption in the unit (21) to be cooled; and a fan output calculation unit (11) that, in regard to the power consumption estimated by the power consumption estimation unit (12), refers to information pertaining to fan output characteristics indicating a fan revolution speed with respect to power consumption and determines the fan output of the cooling fan (23).

Description

Cooling fan control device, cooling fan control method, electronic device, and program

[0001] The present invention relates to a cooling fan control device, a cooling fan control method, an electronic device, and a program.

[0002] Electronic devices and computer systems, such as computers (hereinafter referred to as servers) incorporating CPUs and peripheral devices, have cooling mechanisms to deal with heat generated by semiconductors (chips). The widely used air-cooling method uses a fan to transfer heat to the air, dissipating it outside the enclosure and preventing temperature increases. Conventionally, cooling fan control methods that prevent large changes in the temperature of the cooled component have been known (see Non-Patent Documents 1 and 2). Large changes in the temperature of the cooled component, such as a semiconductor chip such as a CPU, can increase the power consumption of the cooled component or cause it to malfunction.

[0003] M. Zapater, JL Ayala, JM Moya, K. Vaidyanathan, K. Gross, A.K. Coskun, “Leakage and Temperature Aware Server Control for Improving Energy Efficiency in Data Centers,” 2013 Design, Automation & Test in Europe Conference & Exhibition (DATE), pp. 266-269, 2013. Custom Cooling Fan Options for Dell EMC PowerEdge Servers, [online], [Retrieved June 8, 2023], Internet: https: / / dl.dell.com / manuals / common / customcooling_poweredge_idrac9.pdf

[0004] Reducing power consumption is a challenge for electronic devices and computer systems that consume a lot of power. Therefore, it is desirable to minimize the power consumption required for cooling, as long as it does not affect the computing performance for data processing. In addition, versatility is desirable for electronic devices. Versatility means that there is no need to collect information in advance to match the characteristics of the semiconductor chip to be cooled.

[0005] The technology disclosed in Non-Patent Document 1 minimizes power consumption by determining the fan rotation speed so as to minimize the sum of the power consumption increase due to leakage current caused by an increase in CPU temperature and the power consumption by the fan. However, the technology disclosed in Non-Patent Document 1 requires that the correspondence between load conditions and fan output be established in advance for each system, making it unsuitable for general use. Furthermore, the technology disclosed in Non-Patent Document 2 controls the airflow based not only on the object being cooled but also on the intake air temperature, thereby maintaining an airflow that takes temperature fluctuations into account, resulting in increased power consumption. Furthermore, conventional technologies attempt to cool by detecting temperature and rotating the fan when the temperature is high, which requires time from the start of fan control until the heat is released. Furthermore, conventional technologies use feedback control in which temperature fluctuations caused by changing the fan output are used to determine the next fan output. Such feedback control configurations cause oscillations due to positive feedback, which destabilize the temperature and fan rotation speed, requiring countermeasures. As a result, conventional technologies lose versatility and responsiveness and increase power consumption.

[0006] The present invention has been made in view of the above circumstances, and an object of the present invention is to reduce the power consumption required for cooling while maintaining versatility and quick response.

[0007] The cooling fan control device of the present invention is a cooling fan control device that controls a cooling fan, and is characterized by comprising: a power consumption estimation unit that estimates the power consumption of the cooling target unit based on measurement information of the power consumption of the cooling target unit measured by a sensor unit; and a fan output calculation unit that determines the fan output of the cooling fan by referring to information on fan output characteristics that indicate the fan rotation speed relative to the power consumption for the power consumption estimated by the power consumption estimation unit.

[0008] According to the present invention, it is possible to reduce the power consumption required for cooling while maintaining versatility and quick response.

[0009] FIG. 1 is a schematic configuration diagram of a server according to a first embodiment. FIG. 2 is an explanatory diagram of processing by a fan output calculation unit. FIG. 3 is an explanatory diagram of processing by a fail-safe function unit. FIG. 4 is a flowchart showing the flow of cooling fan control processing according to the present embodiment. FIG. 5 is an explanatory diagram of a conventional cooling fan control method. FIG. 6 is an explanatory diagram of a cooling fan control method according to the present embodiment. FIG. 7 is a schematic configuration diagram of a server according to a second embodiment. FIG. 8 is a schematic configuration diagram of a server according to a third embodiment. FIG. 9 is a schematic configuration diagram of a server according to a fourth embodiment. FIG. 10 is a hardware configuration diagram showing an example of a computer that realizes the functions of the server according to the present embodiments.

[0010] A server (electronic device) according to an embodiment will be described in detail below with reference to the drawings. (First Embodiment) [Server Overview] As shown in FIG. 1 , the server 1 includes hardware 20, which includes a cooling target 21, a sensor 22a, a cooling fan 23, and a power supply 24. The sensor 22a measures the load current and power supply voltage of the cooling target 21. The cooling fan 23 cools the cooling target 21. The power supply 24 supplies power to the cooling target 21. The cooling target 21 has a data processing function and is, for example, a CPU (Central Processing Unit). The CPU executes a predetermined application program. The server 1 includes software 10 executed by the CPU. The software 10 is implemented in an operating system (OS) or user space. The software 10 includes a fan output calculation unit 11 and a power consumption estimation unit 12. The power consumption estimation unit 12 estimates the power consumption of the cooling target unit 21 based on measurement information of the power consumption of the cooling target unit 21 measured by the sensor unit 22 a. The fan output calculation unit 11 determines the fan output based on the power consumption estimated by the power consumption estimation unit 12, by referring to information on fan output characteristics that indicate the fan rotation speed relative to the power consumption.

[0011] In this embodiment, the CPU executes a program to realize the cooling fan control device as software 10, and therefore, hereinafter, it may be referred to as the cooling fan control device 10. Each part of the server 1 will be described in detail below.

[0012] [Fan Output Calculation Unit] The fan output calculation unit 11 is provided with information on the power consumption of the cooling target unit 21. Based on the power consumption of the cooling target unit 21, the fan output calculation unit 11 determines a rotation speed corresponding to an air volume capable of cooling the heat generated by that power, and instructs the cooling fan 23. In this embodiment, the fan output calculation unit 11 sets a fan output instruction value in a fan controller (not shown) and changes the fan rotation speed. The fan controller continues to output an instruction for the changed fan rotation speed to the cooling fan 23. Note that the fan controller may be located either external to or internal to the fan output calculation unit 11.

[0013] The fan output calculation unit 11 sets fan output characteristics in advance. The fan output characteristics can be expressed by a relational expression between power consumption and fan rotation speed. Here, as an example, the fan output calculation unit 11 sets a proportionality coefficient (slope) when the fan rotation speed is proportional to power consumption as the fan output characteristics. This fan output calculation unit 11 will be described with reference to FIG. 2A. FIG. 2A is a graph showing the relationship between power consumption and fan output. The horizontal axis represents power consumption, and the vertical axis represents fan output. MIN represents the minimum fan output that can be set for the cooling fan 23. LP represents the upper limit power consumption that can be cooled with the minimum fan output MIN.

[0014] In detail, first, the fan output calculation unit 11 calculates the maximum power consumption (P max ) is the maximum fan output (F max ) indicates the maximum value (P max , F max ) is set. Conversely, a minimum value (LP, MIN) is set that indicates the minimum power consumption (upper limit: LP) that prevents the temperature of the cooling target 21 from exceeding a safe range when the fan rotation speed is kept at the minimum (MIN). Then, the fan output calculation unit 11 calculates the two-dimensional coordinate (P max , Fmax The slope of the line connecting (LP, MIN) and (LP, MIN) (coefficient of power consumption - fan rotation speed) is calculated and set as the fan output characteristic. MIN is set according to the specifications of the cooling fan 23. The upper limit power consumption LP is set by measuring the temperature of the cooling target 21 in advance using the temperature sensor 22b.

[0015] 2A , during idle time when the power consumption of the cooling target 21 is below the upper limit power consumption LP, the fan output calculation unit 11 determines the minimum fan output MIN. On the other hand, when the power consumption of the cooling target 21 is equal to or greater than the upper limit power consumption LP, the fan output calculation unit 11 determines the fan output by proportional control so that the maximum power consumption of the cooling target 21 corresponds to the maximum fan output.

[0016] As a result, the fan output calculation unit 11 can use the estimated power consumption to determine the fan output so as to achieve the highest power efficiency in response to heat generated by the cooling target 21. The most power-efficient fan output can be considered, for example, to be a boundary cooling state that does not reach the fan output (ideal output) that minimizes power consumption. If the fan output (ideal output) is lowered in the above-mentioned ideal output state, cooling will not be sufficient and the temperature of the cooling target 21 will rise. A fan output that cools the cooling target 21 in a cooling state slightly above this fan output (ideal output) corresponds to the most power-efficient fan output.

[0017] [Power Consumption Estimation Unit] The power consumption estimation unit 12 receives information such as load current and power supply voltage from the sensor unit 22a. The power consumption estimation unit 12 calculates or estimates the power consumption of the cooling target 21 from the load current, power supply voltage, and the like. For example, the power consumption estimation unit 12 can estimate the power consumption of the cooling target 21 by reading the load current and power supply voltage measured by the sensor unit 22a and multiplying the load current by the power supply voltage. Power Consumption [W] = Power Supply Voltage [V] × Current [A] The power consumption estimation unit 12 notifies the fan output calculation unit 11 of the power consumption. The power consumption estimation unit 12 can acquire observation data from various sensors in the server 1, and can also acquire the integrated power consumption from the sensor unit 22a and temperature information from the temperature sensor 22b.

[0018] [Fail-safe Function Unit] The cooling fan control device 10 of this embodiment has a fail-safe function unit 13. The fail-safe function unit 13 receives information about the temperature of the cooling target 21 from the temperature sensor 22b. The fail-safe function unit 13 reads the temperature of the cooling target 21 measured by the temperature sensor 22b, and when the temperature exceeds a dangerous temperature threshold, the fail-safe function unit 13 instructs the fan output calculation unit 11 to maximize the fan output. The dangerous temperature threshold is a predetermined temperature threshold (upper limit) that is safe for the cooling target 21. The fail-safe function unit 13 is not essential for the cooling fan control device 10, but it is preferable to have the fail-safe function unit 13 because it can maintain a temperature that does not exceed the dangerous temperature threshold in response to heat generation from the cooling target 21.

[0019] The fail-safe function unit 13 will be described with reference to FIG. 2B. In the graph of FIG. 2B, the horizontal axis represents time and the vertical axis represents temperature. The dashed line represents the critical temperature threshold TH. If the temperature of the cooling target 21 is equal to or higher than the critical temperature threshold TH, the fail-safe function unit 13 sends an instruction to the fan output calculation unit 11 to maximize fan output. In the example shown in FIG. 2B, the temperature of the cooling target 21 is equal to or higher than the critical temperature threshold TH from time T1 to time T2. Therefore, the fan output calculation unit 11 temporarily disables the power consumption-based fan control method from time T1 to time T2 and determines the maximum fan output. This allows the temperature of the cooling target 21 to be quickly reduced. By rapidly cooling the cooling target 21 using the fail-safe function unit 13, the cooling fan control device 10 can prevent breakdowns or failures due to thermal runaway of the cooling target 21. Note that in the example shown in FIG. 2B, the temperature of the cooling target 21 is lower than the critical temperature threshold TH from time T0 to time T1 and after time T2. Therefore, the fan output calculation unit 11 determines the fan output based on the power consumption from time T0 to time T1 and from time T2 onwards.

[0020] [Hardware] The cooling target 21 is a device that is the target of forced air cooling by the cooling fan 23. The cooling target 21 may be a CPU mounted on the motherboard or other devices that consume power. The cooling target 21 may be an expansion device installed outside the motherboard within the housing. The expansion device may be, for example, an accelerator, a network interface card (NIC), a field programmable gate array (FPGA), a graphics processing unit (GPU), an application specific integrated circuit (ASIC), or the like. The number of cooling targets 21 may be one or more.

[0021] The sensor unit 22a acquires the load current, power supply voltage, temperature, air volume, and other information required for fan control from each component, including the cooling target unit 21 and the power supply unit 24, and provides an interface through which this information can be acquired from each functional unit of the software 10. The server 1 of this embodiment also has a temperature sensor 22b and a wind speed sensor (not shown). The temperature sensor 22b measures the temperature of the cooling target unit 21. The cooling target unit 21 may be equipped with at least one of the sensor unit 22a and the temperature sensor 22b. The sensor unit 22a and the temperature sensor 22b may be separate or may be an integrated sensor.

[0022] The cooling fan 23 is a fan that forcibly cools the cooling target 21. The cooling fan 23 is configured so that the rotation speed can be controlled by the fan output calculation unit 11. The number of cooling fans 23 may be one or more. The cooling fan 23 may include, for example, a set of an intake fan and an exhaust fan.

[0023] The power supply unit 24 supplies power to each functional unit of the hardware 20. The sensors for acquiring the power supply voltage, current, and power used for fan control can be provided in the power supply unit 24, as well as in the sensor unit 22a.

[0024] [Operation of Cooling Fan Control Device] Next, the flow of fan control processing by the cooling fan control device 10 will be described with reference to Figure 3 (and also with reference to Figures 1, 2A, and 2B as appropriate). In the cooling fan control device 10, the fan output calculation unit 11 sets the fan output characteristics (step S1). Here, the fan output calculation unit 11 calculates a coefficient of power consumption - fan rotation speed and sets it as the fan output characteristics. Note that the processing of step S1 does not need to be performed each time, and may be performed once in advance.

[0025] In the cooling fan control device 10, the power consumption estimator 12 reads the power supply voltage and load current from the sensor 22a (step S2). The fail-safe function unit 13 acquires the temperature from the temperature sensor 22b. The fail-safe function unit 13 then determines whether the temperature acquired from the temperature sensor 22b is within a safe range (step S3). If the temperature is within the safe range (step S3: Yes), the cooling fan control device 10 performs variable control of the fan rotation speed based on the power consumption. That is, first, the power consumption estimator 12 performs a calculation to estimate the power consumption from the value acquired from the sensor 22a (step S4). Note that this power consumption serves as an index for determining the fan output in the fan output calculation unit 11.

[0026] The fan output calculation unit 11 then calculates the fan output required for cooling from the power consumption estimated by the power consumption estimation unit 12 (step S5). The fan output calculation unit 11 then determines a fan output instruction value based on the fan output characteristics (power consumption - fan rotation speed coefficient). The fan output calculation unit 11 then instructs the cooling fan 23 to output fan power (step S6), and the process returns to step S2. In step S6, the fan output calculation unit 11 sets the fan output instruction value in a fan controller (not shown) and changes the fan rotation speed.

[0027] On the other hand, if the fail-safe function unit 13 determines in step S3 that the temperature acquired from the temperature sensor 22b exceeds the safe range (step S3: No), the cooling fan control device 10 controls the fan output to be maximized. That is, if the temperature of the cooling target 21 is equal to or higher than the dangerous temperature threshold TH, the fail-safe function unit 13 notifies the fan output calculation unit 11 of an instruction to maximize the fan output. The fan output calculation unit 11 then determines to maximize the fan output to prevent thermal runaway (step S7). The fan output calculation unit 11 then proceeds to step S6 and instructs the cooling fan 23 to increase the fan output. At this time, the fan output calculation unit 11 sets the fan output instruction value (maximum value) in a fan controller (not shown) to enable rapid cooling.

[0028] [Effects of Cooling Fan Control Method] Next, the effects of the cooling fan control method of this embodiment will be described in comparison with the conventional technology. First, the conventional cooling fan control method will be described with reference to FIG. 4. In the CPU power consumption graph in the upper part of FIG. 4, the horizontal axis represents time, and the vertical axis represents 0% to 100%. For example, from time t0 to t1, the CPU power consumption is approximately 10%, but from time t1 to t2, it is approximately 30%. Furthermore, from time t2 to t3, the CPU power consumption is approximately 10%, but from time t3 to t4, it is approximately 50%. Furthermore, from time t4 to t5, the CPU power consumption is approximately 10%, but from time t5 to t6, it is approximately 70%.

[0029] In the graph of heat sink temperature in the middle of Figure 4, the horizontal axis represents time and the vertical axis represents temperature. In the graph of heat sink temperature, the dashed line represents the intake air temperature. For example, from time t0 to t1, the heat sink temperature is approximately the same as the intake air temperature, but from time t1 to t2, the heat sink temperature is higher than the intake air temperature. Also, from time t2 to t3, the heat sink temperature is approximately the same as the intake air temperature, but from time t3 to t4, the heat sink temperature is significantly higher than the intake air temperature.

[0030] In the fan output graph at the bottom of Figure 4, the horizontal axis represents time and the vertical axis represents fan output. In the fan output graph, the dashed line represents the fan output lower limit. The solid waveform represents realistic fan output. The two-dot chain waveform represents ideal fan output. In control that changes the fan rotation speed in response to the heat sink temperature, when the intake air temperature is approximately the same as the temperature corresponding to the fan output lower limit, as at times t0 to t1 and t1 to t3, the fan rotation speed decreases, and little power is wasted. On the other hand, when the intake air temperature is higher than the temperature corresponding to the fan output lower limit, as at times t1 to t2 and t3 to t4, the fan rotation speed does not decrease sufficiently, resulting in wasted power consumption.

[0031] When the power consumption of the CPU, which is the part to be cooled, changes, the heat changes, and then the heatsink temperature changes. It also takes time for the heat to be transmitted to the temperature sensor. Therefore, in conventional feedback control using temperature as an index, it takes time for the fan output to change based on the temperature sensor, resulting in insufficient control. For example, if the CPU power consumption suddenly increases at time t5, the fan rotation speed rise time RT is later than time t5. The fan rotation speed command cannot keep up with the temperature fluctuation, and the heatsink temperature rises suddenly from time t5 to the rise time RT. Such a sudden temperature change can lead to increased power consumption and malfunctions.

[0032] Next, a cooling fan control method according to this embodiment will be described with reference to FIG. 5. The CPU power consumption graph in the upper part of FIG. 5 is the same as the CPU power consumption graph in the upper part of FIG. 4. The heat sink temperature graph in the middle part of FIG. 5 is the same as the heat sink temperature graph in the middle part of FIG. 4, except for the temperature change around time t5. In this cooling fan control method, fan output is reduced (within a safe range) regardless of temperature during idle periods such as between times t0 and t1 in the fan output graph in the lower part of FIG. 5, thereby minimizing power consumption.

[0033] Typically, power consumption, heat, temperature, and fan rotation speed change in this order, so conventional cooling fan control methods that use temperature as an index take a long time to control fan output. In contrast, the present cooling fan control method directly determines fan rotation speed based on dynamically changing power consumption, resulting in a quick response. Furthermore, the present cooling fan control method uses power consumption as an index for determining fan output, so it can provide sufficient cooling while determining the most power-efficient fan output and specifying the cooling fan output value. Here, providing sufficient cooling means being able to maintain a temperature that does not exceed a threshold (safe temperature) in response to heat generated by the cooling target 21.

[0034] Power consumption-based control is more responsive and accurate than temperature-based control. Therefore, this cooling fan control method can improve response performance by determining fan output based on power consumption. As shown in Figure 5, if CPU power consumption suddenly increases at time t5, the fan rotation speed (fan output) quickly increases to follow the temperature fluctuation. The heat sink temperature is prevented from rising sharply at time t5, preventing abrupt changes in temperature. This high responsiveness leads to failure prevention. Furthermore, power consumption-based control is more accurate than temperature-based control, minimizing cooling power consumption without affecting computing performance for data processing. Furthermore, this cooling fan control method uses easily obtainable power consumption as an indicator for cooling fan control, making it applicable to many devices. In other words, it is versatile because it does not require prior information collection tailored to the characteristics of the semiconductor chip being cooled.

[0035] Second Embodiment Next, a server according to a second embodiment will be described with reference to FIG. 6 . Components identical to those in the server 1 shown in FIG. 1 are designated by the same reference numerals and will not be described again. The server 1B includes a cooling target 21B. The cooling target 21B has a data processing function and a sensor unit 22a for measuring power consumption. In this embodiment, the CPU (the cooling target 21) has a function for collecting time-varying power consumption information using an internal measurement function. The measured information may be load current and power supply voltage, or integrated power consumption. The software (cooling fan control device) 10B of the server 1B includes a fan output calculation unit 11B and a power consumption estimation unit 12. The power consumption estimation unit 12 estimates the power consumption of the cooling target 21B based on measurement information of the power consumption of the cooling target 21B measured by the sensor unit 22a inside the cooling target 21B. The fan output calculation unit 11B determines the fan output based on the power consumption estimated by the power consumption estimation unit 12.

[0036] The cooling fan control device 10B has a fail-safe function unit 13, an application program (hereinafter referred to as application) 14, and a software index acquisition unit 15. Although these are not essential, it is preferable that the cooling fan control device 10B has these.

[0037] The application 14 is software that runs on the server 1B and provides a desired function. There are no particular limitations on the purpose of the application 14. There may be one or more applications 14. The application 14 can acquire some information and notify the fan output calculation unit 11B of the acquired information via the software index acquisition unit 15, or can notify the fan output calculation unit 11B directly.

[0038] When software index acquisition unit 15 acquires any information from application 14, it notifies fan output calculation unit 11 of the information as is, or after processing it as necessary. Software index acquisition unit 15 can perform calculations using a predetermined algorithm based on the information collected from one or more applications 14, and notify fan output calculation unit 11 of the calculation results.

[0039] The application 14 can also acquire information about power consumption sensed inside the cooling target 21B. In this case, the fan output calculation unit 11B determines the fan output and issues instructions to the cooling fan 23 using the power consumption information acquired by the application 14, giving priority to the estimated power consumption.

[0040] When the application 14 acquires information on the load current and the power supply voltage from the sensor unit 22a, the application 14 multiplies the load current by the power supply voltage to obtain information on the power consumption. Alternatively, the software indicator acquisition unit 15 multiplies the load current by the power supply voltage to obtain information on the power consumption. The application 14 can also acquire the power consumption as an integrated result from the sensor unit 22a.

[0041] Third Embodiment Next, a server according to a third embodiment will be described with reference to Fig. 7. Components identical to those of the server 1 shown in Fig. 1 and the server 1B shown in Fig. 6 are designated by the same reference numerals, and descriptions thereof will be omitted. As shown in Fig. 7, the server 1C includes, as hardware 20, a cooling target unit 21, a sensor unit 22a, a temperature sensor 22b, a cooling fan 23, and a power supply unit 24. The software (cooling fan control device) 10C of the server 1C includes a fan output calculation unit 11C, a power consumption estimation unit 12, a fail-safe function unit 13, an application 14, and a software index acquisition unit 15.

[0042] The cooling fan control device 10C normally controls the cooling fan 23 based on the power consumption measured by the sensor unit 22a, but if there is an index that causes a temperature rise earlier than the power consumption, the index is used for fan control. If there is a load leading index that causes a temperature rise before the power consumption rises, the application 14 notifies the software index acquisition unit 15 of the load leading index.

[0043] If the application 14 has a leading load indicator, the software indicator acquisition unit 15 acquires the leading load indicator, calculates prediction information that predicts the power consumption or heat generation of the cooling target unit 21 from the leading load indicator, and notifies the fan output calculation unit 11C of the prediction information. When the fan output calculation unit 11C acquires the prediction information, it determines the fan output using the prediction information, giving priority to the estimated power consumption.

[0044] The load leading indicator is an indicator of load fluctuations on the cooling target unit 21 and is application-specific information. For example, if the data to be processed by the application 14 is traffic, traffic fluctuations and the like become the load leading indicator. Due to traffic fluctuations and the like, the load on the CPU (cooling target unit 21) executing the application 14 fluctuates over time. If the load leading indicator indicates that the CPU is under high load, the fan output can be increased to suppress an increase in the CPU temperature. Therefore, the fan output calculation unit 11C can precisely control the fan output in accordance with the current load situation. Note that the load leading indicator may not always be present. If the load leading indicator cannot be acquired, the cooling fan control device 10C determines the fan output based on the sensed power consumption, similar to the cooling fan control device 10 according to the first embodiment.

[0045] Fourth Embodiment Next, a server according to a fourth embodiment will be described with reference to Fig. 8. Components identical to those of the server 1C shown in Fig. 7 are designated by the same reference numerals, and description thereof will be omitted. As shown in Fig. 8, a server 1D includes, as hardware 20, a cooling target unit 21, a sensor unit 22a, a temperature sensor 22b, a cooling fan 23, and a power supply unit 24. Software (cooling fan control device) 10D of the server 1D includes a fan output calculation unit 11D, a power consumption estimation unit 12D, a fail-safe function unit 13, an application 14, and a software index acquisition unit 15D.

[0046] There are two differences between the cooling fan control device 10D and the cooling fan control device 10C. The first difference is that, when the application 14 has a load leading indicator, the software indicator acquisition unit 15D acquires the load leading indicator and notifies the power consumption estimator 12D of prediction information that predicts the power consumption or heat generation of the cooling target unit 21 from the load leading indicator. The second difference is that, when the power consumption estimator 12D acquires prediction information, it uses the prediction information in priority over information measured by the sensor unit 22a to estimate the power consumption of the cooling target unit 21.

[0047] Like the fan output calculation unit 11C, the fan output calculation unit 11D can precisely control the fan output in accordance with the current load situation. The power consumption estimation unit 12D can also directly acquire the load leading indicator from the software indicator acquisition unit 15D and calculate or estimate the power consumption of the cooling target unit 21 from the leading indicator. Note that if the load leading indicator cannot be acquired, the cooling fan control device 10D determines the fan output based on the sensed power consumption, similar to the cooling fan control device 10 according to the first embodiment.

[0048] [Hardware Configuration] The server 1 and cooling fan control device 10 according to the embodiment are realized by a computer 900 configured as shown in Fig. 9. The computer 900 includes a CPU 901, a RAM (Random Access Memory) 902, a ROM (Read Only Memory) 903, an HDD (Hard Disk Drive) 904, an accelerator 905, an input / output I / F (Interface) 906, a media I / F 907, and a communication I / F 908. The CPU 901 and the accelerator 905 correspond to the cooling target 21 of the server 1 shown in Fig. 1.

[0049] The accelerator 905 processes at high speed at least one of data from the communication I / F 908 and data from the RAM 902. Note that the accelerator 905 may be of a type (look-aside type) that executes processing from the CPU 901 or RAM 902 and then returns the execution results to the CPU 901 or RAM 902. Alternatively, the accelerator 905 may be of a type (in-line type) that performs processing between the communication I / F 908 and the CPU 901 or RAM 902.

[0050] The accelerator 905 is connected to an external device 915 via a communication I / F 908. The input / output I / F 906 is connected to an input / output device 916. The media I / F 907 reads and writes data from and to a recording medium 917.

[0051] The CPU 901 operates based on a program stored in the ROM 903 or HDD 904, and controls each part of the server 1 shown in Fig. 1 by executing a program loaded into the RAM 902. This program can also be distributed via a communication line or recorded on a recording medium 917 such as a CD-ROM. The ROM 903 stores a boot program executed by the CPU 901 when the computer 900 starts up, programs related to the hardware of the computer 900, and the like.

[0052] The CPU 901 controls an input / output device 916, which is made up of input units such as a mouse and a keyboard, and output units such as a display and a printer, via an input / output I / F 906. The CPU 901 acquires data from the input / output device 916 via the input / output I / F 906, and outputs generated data to the input / output device 916. Note that a GPU or the like can also be used as a processor together with the CPU 901.

[0053] The HDD 904 stores programs executed by the CPU 901 and data used by the programs. The communication I / F 908 receives data from other devices via a communication network and outputs the data to the CPU 901, and also transmits data generated by the CPU 901 to other devices via the communication network.

[0054] The media I / F 907 reads a program or data stored in a recording medium 917 and outputs it to the CPU 901 via the RAM 902. The CPU 901 loads a program related to a target process from the recording medium 917 onto the RAM 902 via the media I / F 907, and executes the loaded program. The recording medium 917 is an optical recording medium such as a DVD (Digital Versatile Disc) or a PD (Phase Change Rewritable Disc), a magneto-optical recording medium such as an MO (Magneto Optical Disk), a magnetic recording medium, or a semiconductor memory.

[0055] For example, when the computer 900 functions as the server 1 according to the embodiment, the CPU 901 executes a program loaded onto the RAM 902 to realize the functions of the server 1. For example, when the computer 900 functions as the cooling fan control device 10 according to the embodiment, the CPU 901 executes a program loaded onto the RAM 902 to realize the functions of the cooling fan control device 10. In addition, the HDD 904 stores data in the RAM 902. The CPU 901 reads and executes a program related to a target process from the recording medium 917. Alternatively, the CPU 901 may read a program related to a target process from another device via a communication network.

[0056] [Effects] As described above, the cooling fan control device is the cooling fan control device 10 that controls the cooling fan 23, and is characterized by including: a power consumption estimation unit 12 that estimates the power consumption of the cooling target unit 21 based on measurement information of the power consumption in the cooling target unit 21 measured by the sensor unit 22 a; and a fan output calculation unit 11 that determines the fan output of the cooling fan 23 by referring to information on fan output characteristics that indicate the fan rotation speed relative to the power consumption for the power consumption estimated by the power consumption estimation unit 12.

[0057] In this way, the cooling fan control device 10 determines the fan output of the cooling fan 23 using power consumption, thereby providing a more responsive system than temperature-based fan control. Because the cooling fan control device 10 has a responsive system, it can effectively prevent malfunctions. Furthermore, because control based on power consumption is more accurate than control based on temperature, the cooling fan control device 10 can minimize the power consumption required for cooling without affecting the computing performance for data processing. Furthermore, because the cooling fan control device 10 uses dynamically changing power consumption as an indicator for cooling fan control, there is no need to collect information in advance, making it versatile.

[0058] The cooling fan control method is a cooling fan control method of a cooling fan control device 10, and is characterized in that the cooling fan control device 10 executes the steps of: estimating the power consumption of the cooling target part 21 based on measurement information of the power consumption in the cooling target part 21 measured by a sensor part 22a; and determining the fan output of the cooling fan 23 based on the estimated power consumption by referring to information on fan output characteristics that indicate the fan rotation speed relative to the power consumption.

[0059] In this way, in the cooling fan control method, the cooling fan controller 10 determines the fan output of the cooling fan 23 using power consumption, thereby improving responsiveness compared to temperature-based control. Furthermore, because power consumption-based control is more accurate than temperature-based control, the cooling fan controller 10 can minimize the power consumption required for cooling without affecting the computing performance for data processing. Furthermore, because the cooling fan controller 10 uses dynamically changing power consumption as an indicator for cooling fan control, there is no need to collect information in advance, making it versatile.

[0060] The electronic device 1 has a CPU (Central Processing Unit) that executes an application (application program) 14, a cooling target 21 that has a data processing function, a power supply unit 24 that supplies power to the cooling target 21, a cooling fan 23 that cools the cooling target 21, a sensor unit 22a that measures the load current and power supply voltage of the cooling target 21, and software 10 that is implemented in an OS (Operating System) or user space and executed by the CPU, and the software 10 is characterized by having a power consumption estimation unit 12 that estimates the power consumption of the cooling target 21 based on measurement information of the power consumption of the cooling target 21 measured by the sensor unit 22a, and a fan output calculation unit 11 that determines the fan output by referring to information on fan output characteristics that indicate the fan rotation speed relative to the power consumption for the power consumption estimated by the power consumption estimation unit 12.

[0061] In this way, the electronic device 1 can reduce construction costs compared to when a cooling fan control function is constructed using dedicated hardware by equipping hardware with a CPU with the software 10 that determines the fan output of the cooling fan 23 using power consumption. In the electronic device 1, the software 10 determines the fan output of the cooling fan 23 using power consumption, thereby improving responsiveness compared to temperature-based control.

[0062] The electronic device 1B has a CPU that executes an application 14, a cooling target part 21B that has a data processing function and a sensor part 22a that measures power consumption, a power supply part 24 that supplies power to the cooling target part 21B, a cooling fan 23 that cools the cooling target part 21B, and software 10B that is implemented in an OS or user space and executed by the CPU, and the software 10B is characterized by having a power consumption estimation part 12 that estimates the power consumption of the cooling target part 21B based on measurement information of the power consumption of the cooling target part 21B measured by the sensor part 22a, and a fan output calculation part 11B that determines the fan output by referring to information on fan output characteristics that indicate the fan rotation speed relative to the power consumption for the power consumption estimated by the power consumption estimation part 12.

[0063] In this way, electronic device 1B is provided with cooling target 21B having a data processing function and sensor unit 22a for measuring power consumption, so there is no need to provide a sensor unit for measuring power consumption separately from cooling target 21B, thereby reducing construction costs. In electronic device 1B, software 10B determines the fan output of cooling fan 23 using power consumption, thereby improving responsiveness compared to temperature-based control.

[0064] In electronic device 1B, when application 14 acquires information on power consumption sensed inside cooling target part 21B, fan output calculation part 11B determines the fan output using the power consumption information acquired by application 14, giving priority to the estimated power consumption.

[0065] By doing this, in electronic device 1B, when application 14 acquires information on power consumption sensed inside cooling target part 21B, fan output calculation part 11B can determine the fan output early without waiting for the power consumption calculation result by power consumption estimation part 12.

[0066] In the electronic device 1C, the software 10C has a software index acquisition unit 15 that acquires the load leading index, which is an index that causes a temperature rise before an increase in power consumption, when the application 14 has the load leading index, calculates prediction information that predicts the power consumption or heat generation of the cooling target part 21 from the load leading index, and notifies the fan output calculation unit 11C of the prediction information, and when the fan output calculation unit 11C acquires the prediction information, it uses the prediction information in priority to the estimated power consumption to determine the fan output.

[0067] By doing this, when a load leading indicator is present in the electronic device 1C, the fan output calculation unit 11C can increase the fan output faster than control based on power consumption, thereby suppressing a rise in CPU temperature.

[0068] In the electronic device 1D, the software 10D has a software index acquisition unit 15D that acquires a load leading index, which is an index that causes a temperature increase before an increase in power consumption, when the application 14 has the load leading index, and notifies the power consumption estimation unit 12D of prediction information that predicts the power consumption or heat generation of the cooling target unit 21 from the load leading index, and when the power consumption estimation unit 12D acquires the prediction information, it uses the prediction information in priority to the information measured by the sensor unit 22a to estimate the power consumption of the cooling target unit 21.

[0069] By doing this, when a load leading indicator is present in the electronic device 1D, the power consumption estimation unit 12D can estimate power consumption more quickly and accurately by taking into account prediction information based on the load leading indicator.

[0070] The present invention is not limited to the above-described embodiments, and many modifications within the scope of the present invention are possible by those skilled in the art. For example, the electronic device is not limited to a computer (server) as long as it has a cooling target and a cooling fan that receives power and performs data processing. The electronic device may be, for example, a personal computer or a home appliance incorporating a CPU and peripheral devices.

[0071] DESCRIPTION OF SYMBOLS 1, 1B, 1C, 1D Server (electronic device) 10, 10B, 10C, 10D Software (cooling fan control device) 11, 11B, 11C, 11D Fan output calculation unit 12, 12D Power consumption estimation unit 13 Fail-safe function unit 14 Application (application program) 15, 15D Software index acquisition unit 20 Hardware 21, 21B Cooling target unit 22a Sensor unit 22b Temperature sensor 23 Cooling fan 24 Power supply unit

Claims

1. A cooling fan control device that controls a cooling fan, comprising: a power consumption estimation unit that estimates the power consumption of a part to be cooled based on measurement information of the power consumption of the part to be cooled measured by a sensor unit; and a fan output calculation unit that determines the fan output of the cooling fan by referring to information on fan output characteristics that indicate the fan rotation speed relative to the power consumption for the power consumption estimated by the power consumption estimation unit.

2. A cooling fan control method for a cooling fan control device, wherein the cooling fan control device executes the steps of: estimating the power consumption of the object to be cooled based on measurement information of the power consumption of the object to be cooled measured by a sensor unit; and determining the fan output of the cooling fan by referring to information on fan output characteristics indicating the fan rotation speed relative to the power consumption for the estimated power consumption.

3. An electronic device having a CPU (Central Processing Unit) that executes an application program, a cooling target having a data processing function, a power supply unit that supplies power to the cooling target, a cooling fan that cools the cooling target, a sensor unit that measures the load current and power supply voltage of the cooling target, and software that is implemented in an OS (Operating System) or user space and executed by the CPU, wherein the software has: a power consumption estimation unit that estimates the power consumption of the cooling target based on measurement information of the power consumption of the cooling target measured by the sensor unit, and a fan output calculation unit that determines fan output by referring to information on fan output characteristics that indicate the fan rotation speed relative to the power consumption for the power consumption estimated by the power consumption estimation unit.

4. An electronic device comprising: a CPU for executing an application program; a cooling target part having a data processing function and a sensor part for measuring power consumption; a power supply part for supplying power to the cooling target part; a cooling fan for cooling the cooling target part; and software implemented in an OS or user space and executed by the CPU, wherein the software comprises: a power consumption estimation part for estimating the power consumption of the cooling target part based on measurement information of the power consumption of the cooling target part measured by the sensor part; and a fan output calculation part for determining fan output by referring to information on fan output characteristics indicating the fan rotation speed relative to the power consumption for the power consumption estimated by the power consumption estimation part.

5. The electronic device described in claim 4, characterized in that when the application program acquires information on power consumption sensed inside the cooling target, the fan output calculation unit determines the fan output using the power consumption information acquired by the application program, giving priority to estimated power consumption.

6. The electronic device described in any one of claims 3 to 5, characterized in that the software has a software index acquisition unit that, when the application program contains a load leading indicator which is an indicator that causes a temperature rise prior to an increase in power consumption, acquires the load leading indicator, calculates prediction information that predicts the power consumption or heat generation of the cooling target from the load leading indicator, and notifies the fan output calculation unit of the prediction information, when the fan output calculation unit acquires the prediction information, using the prediction information in priority to estimated power consumption to determine the fan output.

7. The software has a software indicator acquisition unit that, when a load leading indicator, which is an indicator that causes a temperature increase prior to an increase in power consumption, acquires the load leading indicator when the application program exists, and notifies the power consumption estimation unit of prediction information that predicts the power consumption or heat generation of the cooling target part from the load leading indicator; and the power consumption estimation unit, when it acquires the prediction information, estimates the power consumption of the cooling target part using the prediction information in priority to information measured by the sensor unit, characterized in that the electronic device described in any one of claims 3 to 5.

8. A program for causing a computer to function as the cooling fan control device according to claim 1.