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

Definitions

• the present invention relates to heat generation control technology, and more particularly to a method and apparatus for controlling heat generation of hardware in a computer system.
• the present invention has been made in view of the above circumstances, and an object thereof is to provide a technique for efficiently controlling heat generation of hardware in a computer system.
• An aspect of the present invention relates to a method of controlling heat generation of a target portion to be controlled in a heat generation state in hardware of a computer system.
• This method is The heat generation state of the target portion is controlled by acquiring the heat generation state and changing the operation content of the operating application so as to change the heat generation state according to the acquired heat generation state.
• the present invention as a storage medium storing an apparatus, a system, a program, and a program is also effective as an aspect of the present invention.
• the present invention is advantageous in controlling the heat generation of hardware in a computer system.
• FIG. 1 is a diagram showing an example of a control method list.
• FIG. 2 is a diagram showing a computer system according to an embodiment of the present invention.
• FIG. 3 is a view showing the configuration of a software stack in the computer system shown in FIG. 2;
• FIG. 4 is a diagram showing an example of a reaction table that defines a control method.
• FIG. 5 is a flowchart showing processing of a control execution unit in the control stack shown in FIG.
• FIG. 6 A flowchart showing hardware control processing for adjusting hardware operation.
• FIG. 7 A flowchart showing software control processing for changing the operation content of an application.
• Computer system caused by temperature rise inside device using computer system In the event of problems such as a malfunction, a hang-up, or abnormal termination, for example, increase the fins of the heat sink or increase the power of the cooling fan to increase the cooling capacity of the cooling mechanism to prevent the temperature rise itself. Conceivable.
• a game machine will be considered as an example.
• the maker of a game machine is usually different from the maker of an application operating on the game machine, that is, game software (hereinafter referred to as game software).
• game software hereinafter referred to as game software
• the heat generation amount of the game machine hardware changes while this game software is operating.
• the amount of heat generation during operation of the game machine varies depending on the specifications of the game software in operation and the usage state such as the number of functions being used simultaneously.
• the game machine manufacturer will design a game machine with a margin for the cooling capacity of the cooling mechanism!
• This method has a problem that the size or power consumption of the game machine is increased. Also, in actual use, when the hardware of the game console does not generate so much heat as it causes trouble in the system, the cooling capacity of the cooling mechanism is often wasted compared to the actual calorific value.
• the present inventor proposes the following technology in order to control heat generation of the hardware of the computer system efficiently.
• This technology acquires the heat generation state of the target portion to be controlled in the heat generation state in the hardware of the computer system, and changes the heat generation state according to the acquired heat generation state.
• the operation content of the active application is changed.
• change the operation content of the application means to directly operate the application and change the operation content so that the processing capability of the hardware required for the application changes. And this change may change the heat generation status of the target site.
• the operating mode of an application operating in the normal mode may be changed to a mode slower than the processing speed required by the normal mode, to a slow mode requiring the processing speed, or the main functional parts of the abbr Leave mode and change the processing mode of other functions to slow mode. Stopping the running application is also included in this "modification".
• the operation of a part included in the hardware of the system and whose operation state affects the heat generation state of the target part may be adjusted.
• adjustment means directly adjusting the operation of the above-mentioned part, and for example, adjusting the operating frequency of the processor, adjusting the rotational speed of the fan, etc. Can.
• the heat generation state of the target portion may be controlled by using both the change of the operation content of the application and the operation adjustment of the hardware.
• control method list that defines the control method for each heat generation state that can be taken by the target part is held, and the control method corresponding to the heat generation state of the target part is acquired from this control method list. Let's try to control the heat generation state of the target part by the control method.
• control method indicates a method of changing the operation content of the application, a method of adjusting the hardware, or a combination thereof.
• control method list will be specifically described using an example of a system that has a main processor and a graphic processor and can execute the game function and the PVR function.
• the game function and the PVR function are realized by the game LPAR and PVRLPAR.
• An LPAR is a logical partition function or logical partition function that logically divides and uses system resources, and in this specification is the same as a stack of software executed on a logically divided partition. With the meaning The software belonging to the same LPAR is of the same type.
• FIG. 1 shows an example of a control method list during operation of a game LP AR out of two LPARs in this system.
• the levels of heat generation in the main processor and graphic processor are divided into nine levels from A to I in descending order of temperature, and the control method is determined according to each level.
• main processor operation suppression is determined as the control method "9".
• Main processor operation suppression here is direct adjustment of hardware operation such as lowering the frequency of the main processor.
• the control method “8 & 11 & 12” is “game on top of data protection LP AR suspend, warning message display, log recording” Is decided.
• the “game LPAR suspend on data protection” is a change in the operation content of the software, and this control puts the game LPAR in the suspend mode by giving a grace period such as saving data, for example.
• the highest level A indicates a dangerous state such as immediately before the main processor can not operate normally.
• system shutdown is determined as the control method “1”.
• a control method is determined according to the level of the heat generation state.
• a priority is given to each type of application, and the operation content is changed so as to suppress the operation of the application of a type with lower priority according to the priority of the type of application in operation.
• each function implemented by the application is given a priority, and the operation content is controlled so as to suppress the operation of the part implementing this function as the function with lower priority with respect to the application running. Let me change it.
• the heat generation state of the target site when the heat generation state of the target site is out of the specified normal range, the heat generation state of the target site is changed back to the normal state range. Or, after adjustment, after this change or adjustment, it is preferable to perform return processing, that is, cancellation of the change, restoration of adjustment, when the heat generation state of the target site returns to the normal state range.
• the user may be notified when the heat generation state of the target site deviates from the specified normal state range.
• the frequency at which the heat generation state of the target site deviates from the normal state is constant. It is preferable to notify only when you exceed the threshold U ,.
• FIG. 2 is a block diagram showing a configuration of a computer system 500 according to an embodiment of the present invention.
• the computer system 500 is used as an example in a game machine, and this game machine has a function as a PVR in addition to the game function.
• Computer system 500 provides functions and an environment for efficiently using computer system 500, and an operating system (hereinafter referred to as an OS) that centrally controls the entire system is executed. Multiple application software (hereinafter referred to as application) are executed on the OS.
• an operating system hereinafter referred to as an OS
• application Multiple application software
• each element described as a functional block that performs various processing can be configured as a CPU, a memory, and other LSIs in terms of hardware, and in software terms It is realized by a program with reservation management function loaded in memory. Therefore, it is understood by those skilled in the art that these functional blocks can be realized in various forms by hardware only, software only, or a combination thereof, and is not limited to any of them.
• Computer system 500 includes a system board 100 and a software stack 200.
• the system board 100 includes a main processor 10, a graphic processor 20, and a transmitting unit 40.
• the main processor 10 and the graphic processor 20 cooperate to perform arithmetic processing Wow.
• Temperature sensors 30 are provided inside the chips of the main processor 10 and the graphic processor 20, respectively.
• the transmitter 40 connected to the temperature sensor 30 transmits the temperature of the main processor 10 and the graphic processor 20 measured by the temperature sensor 30 to the software stack 200, specifically, the control stack 110 described later.
• Software stack 200 includes control stack 110, system LPAR 120, utility L PARI 30, game LPAR 140, and PVRL PAR 150.
• the system LPAR 120 is an LP AR that implements the basic functions of the OS of the computer system 500, and the utility LPAR 130 supports a display device (not shown) and drivers for various peripheral devices.
• the games LPAR 140 and PVRL PAR 150 are LPARs that execute game functions and PVR functions as application LPARs.
• Control stack 110 controls LPARs, such as scheduling of each LPAR included in software stack 200, and also controls the operation of hardware included in computer system 500, in this case, each configuration included in system board 100. It can also be done.
• control of the heat generation state of the system board 100 is performed by the control stack 110.
• FIG. 3 illustrates the functionality of control stack 110. In order to facilitate understanding of the present invention, FIG. 3 shows the functions involved in controlling the heat generation of the system board 100 among the functions executed by the control stack 110, and the other functions will be described. Is omitted here.
• FIG. 3 shows a control stack 110.
• the control stack 110 includes a receiver 112, a control execution unit 114, a type acquisition unit 116, and a reaction table 118.
• the receiving unit 112 receives the temperatures of the main processor 10 and the graphic processor 20 transmitted from the transmitting unit 40, the type acquisition unit 116 acquires the type of the operating application, and the control execution unit 114 Control is performed according to the temperature received by the receiving unit 112 and the type of the operating application acquired by the type acquisition unit 116.
• the reaction table 118 is a list that defines the control method for the application in the control performed by the control execution unit 114.
• the control execution unit 114 transmits the temperature received by the reception unit 112 and the type acquisition unit 116 from the reaction table 118.
• the control method corresponding to the type of the application in operation acquired from the acquired is acquired, and the application is controlled by the acquired control method.
• control execution unit 114 will be specifically described with reference to FIGS. 4 to 7.
• FIG. 4 shows a specific example of the reaction table 118.
• the reaction table 118 may be implemented in any form as long as the control execution unit 114 can refer to it.
• the reaction table 118 is implemented as a library.
• FIG. 5 to 7 are flowcharts showing the processing of the control execution unit 114.
• the control execution unit 114 When performing control, the control execution unit 114 counts the number i of thermal error occurrences, and the default value of the number i is 0 (S10).
• the control execution unit 114 assumes that a thermal error has occurred when the temperature of any of the main processor 10 and the graphic processor 20 exceeds a predetermined threshold, and starts control (S 14).
• the control execution unit 114 first confirms the number i of occurrences of the error (S16).
• the control execution unit 114 determines the warning reference time T1 and the hardware control reference time T2 (for details of the warning reference time T1 and the hardware control reference time ⁇ 2).
• the current time t is set as will be described later, and 1 is added to the number i of thermal error occurrences (S40, S44). Then, the control execution unit 114 performs hardware control processing (S50).
• FIG. 6 is a flowchart showing the hardware control process of step S50.
• the hardware control processing performed by the control execution unit 114 is included in the hardware of the computer system 500, and the operation of the part that affects the heat generation state of the target part whose operation is the control target of the heat generation state is directly adjusted.
• the target portion is the main processor 10 and the graphic processor 20, and the control execution unit 114 sets the portion to be adjusted as the target portion itself.
• control execution unit 114 In executing hardware control processing, control execution unit 114 first confirms which processor temperature has exceeded the threshold (S 51, S 54). 0 Both main processor 10 and graphic processor 20 have temperatures When the threshold value is exceeded (S51: Yes), the control execution unit 114 reduces the power consumption of the main processor 10 and the graphic processor 20 by using a method of reducing the operating frequency of the processor such as clock gating, for example. Both operations are suppressed (S52).
• a method of suppressing the operation of the main processor 10 and the graphic processor 20 the same one may be used for the main processor 10 and the graphic processor 20, or different methods corresponding to respective architectures may be used. It may be
• the control execution unit 114 suppresses the operation of the main processor 10 (S56).
• the control execution unit 114 suppresses the operation of the graphic processor 20 (S58).
• the heat generation state of the graphic processor 20 may be affected. Therefore, when the operation of the graphic processor 20 is suppressed, the operation of the main processor 10 may be suppressed as well. ,.
• the control execution unit 114 returns when the temperature of the target site returns to below the threshold (S70: Yes, S60: Yes).
• the process is performed (S80), and the process waits until the next error occurs.
• the recovery process here is a process of restoring the operation of the target part subjected to the adjustment in step S50, and may be, for example, a process of returning the operating frequency of the processor to the original frequency.
• the control execution unit 114 performs the software control process. Do (S 100).
• step S16 if it is not the number of occurrences of an error (S16: No), the control execution unit 114 determines from the warning reference time T1 set in step S40 to the current error occurrence time. The time (t T1) is confirmed (S20). If this time exceeds the predetermined threshold A (S20: Yes), the control execution unit 114 resets the error occurrence number i to 0 and sets the warning reference time T1 to t at the current error occurrence. (S24: S40), the processing from step S44 is performed. On the other hand, when the time (t ⁇ T1) is within the threshold A (S20: No), the control execution unit 114 calculates the occurrence frequency of the error within this time, and the calculated occurrence frequency is the threshold B. It is checked whether or not the car is exceeded (S30).
• step S44 has already been described, so the description is omitted here.
• FIG. 7 is a flowchart showing the software control process of step S100 in the flowchart shown in FIG.
• the control execution unit 114 performs software control processing when the temperature of the target portion can not be returned to the threshold value or less within the time limit Ch. This software control process is performed with reference to the reaction table 118 shown in FIG.
• the reaction table 118 is a list that gradually defines a control method (reaction in the figure) for each combination of the processor in which the thermal error has occurred and the type of application in operation.
• reaction table 118 and the software control process will be described in detail, taking the case where the thermal error occurs only in the main processor 10 and the type of the active application is only a game.
• the control execution unit 114 first causes the type acquisition unit 116 to acquire the type of the application LPAR in operation (S104).
• the type of application LPAR in operation is the game LPAR 140.
• the control execution unit 114 acquires corresponding control methods from the reaction table 118 based on the processor in which the thermal error has occurred and the type of the application LP AR in operation (S 108). As shown in the top column of Fig. 4, the control methods for this case are "silent mode”, “silent & low resolution mode”, "game suspend”, “game shutdown”, “system suspend” and “system shutdown”. The order is defined in stages. The control execution unit 114 performs control in this order.
• the game LPAR 140 is defined to suppress its operation. Be done.
• the image drawing function of the game application has higher priority than the voice function, and the first low priority voice function is suppressed when the operation of the game application is suppressed, the corresponding column of the reaction table 118 is selected.
• "Silent mode" is defined as the control method of the first step!
• the control execution unit 114 first sets the software control reference time T3 to the current time, and first stops the voice function of the game application according to the control method “silent mode” of the first step (S 110 , S114: No, SI 20, SI 24). Then, when the temperature of the main processor 10 returns below the threshold value within the predetermined time limit Cs from the time when the voice function is stopped, that is, the software control reference time T3, (S130: Yes, S12 8 Yes: Perform recovery processing (S80 in Fig. 5) and wait until the next error occurs.
• the return process in the software control process is a process of canceling the change made in step S124, and is a process of recovering the voice function of the game application.
• the control execution unit 114 performs the processing from S 120 according to the control method of the next step while continuing the control already performed.
• the control method “low resolution mode” of the second step is used, the silent mode is continued, and the image drawing mode of the game application is set to the low resolution mode.
• control execution unit 114 sequentially uses a plurality of defined control methods and the control by a certain control method returns the temperature of the target site to the threshold or less within the time limit Cs.
• the temperature of the target site can not be returned to the threshold or less within the control power time limit Cs by a certain control method. Control is performed by adding the following control method.
• control execution unit 114 shuts down the entire computer system 500 (S150).
• reaction table 118 In the top column of reaction table 118, the third, fourth, fifth, and sixth steps are controlled.
• control method "game suspend”, “game shutdown”, “system suspend” and “system shutdown” are defined, and the control operation corresponding to each is "game
• the following operations are performed: put the LPAR 140 in suspend mode, shut down the game LPAR 140, put the entire system in suspend mode, and shut down the entire system.
• control execution unit 114 secures a grace period for saving data and saves data, when putting the application LPAR or the entire system into the suspend mode. Then go into suspend mode
• the thermal error power occurs only in the main processor 10, and the type of application being operated is PVR only, compared with "PVR slow mode", “PVR suspend”, "
• a five-step control method is defined in the following order: PVR shutdown, system suspend, and system shutdown! “PVR slow mode”, “PVR suspend”, and “PVR shutdown” correspond to the control actions “set recording mode to slow mode”, “put PVRLPAR 150 to suspend mode”, and “shut down PVRL PAR 150”, respectively. .
• the thermal error occurs only during the operation of this low priority application.
• the control method is defined in the order of “PVR slow mode”, “PVR suspend”, and “PVR shutdown”. ing.
• the PVR slow mode is selected for PVR to be suppressed first. Control methods in the order of “PVR suspend” and “PVR shutdown” may be used.
• the load on the processor is set higher than the control method of “PVR low speed mode” which is not “PVR low speed mode” as the control method of the first step. It defines “PVR suspension bends” that can be mitigated. By doing this, the impact on high priority applications can be eliminated quickly.
• the type of the active application is the game only Similarly to the above, a control method of five steps is defined in the order of “game low resolution mode”, “game suspend”, “game shutdown”, “system suspend” and “system shutdown”.
• control methods are specified in the order of These six control methods are control methods when thermal errors occur simultaneously in the main processor 10 and the graphic processor 20, and the main processor 10 or the graphic processor is controlled during control by any of the control methods. If the 20 thermal errors are eliminated, the control execution unit 114 determines from the control method of the step that a minor error is eliminated and a control method in the case where a minor error occurs only in a weak processor.
• Control with For example, when control is performed in the control method “PVR shut down, game low resolution mode” in the first step, and the temperature of the main processor 10 falls below the threshold within the time limit Cs, the control execution unit 114 Of the control methods specified for “When the thermal error occurs in the graphic processor only and the running application is a game”, control is performed using the “game suspend” control method next to “Low-resolution mode”. Do.
• the computer system 500 shown in FIG. 2 measures the temperatures of the main processor 10 and the graphic processor 20 to be controlled in the heat generation state, and transmits them to the control stack 110, and the control stack 110 controls Control the temperature of the control target by changing the operation content of the application according to the target temperature. By doing this, reliable temperature control can be realized without providing a system with a large power cooling device.
• the image drawing function of the game function is given higher priority than the audio function, and when the operation is suppressed, the functional power with low priority is also suppressed. It is possible to continue to execute high-priority functions while reducing the stress on the user.
• priority is given to the application LPAR to suppress the operation from the low priority!, LPAR. By doing this, it is possible to continue to execute the main functions of the system, that is, high priority applications, while suppressing the temperature. Furthermore, temperature control is realized more efficiently by simultaneously adjusting the operation of the main processor and graphic processor, such as reducing the operating frequency of the main processor and the graphic processor.
• reaction table 118 defining the control method is provided for each possible heating condition of the system, control execution unit 114 can perform control with reference to this table, and the configuration is complicated. Systems can easily be controlled.
• the user since the user is warned of the occurrence of a thermal error, the user can check whether the operating environment has a problem or not according to it, which is safe. . In normal use, it is assumed that thermal errors are unlikely to occur frequently, and the user is warned by the frequent alert by alerting only when the frequency of thermal error exceeds a predetermined threshold. The user can be informed that there may be a problem in the usage environment, etc., as well as avoiding the feeling of being bothersome.
• the computer system 500 shown in FIG. 2 is a system used for a game machine
• the heat generation control technology according to the present invention is not limited to a game machine, and a computer system is also used. It can apply.
• the type of application is divided into two types of game and PVR.
• the type of application is not limited to two, and may be the following.
• the thermal error occurrence history may be recorded as a log. At that time, it is desirable to record the location of the thermal error and the type of application in operation. Later, based on the logs, it is possible to analyze the heat generation factor of the system in detail, and it can be used for design improvement.
• the occurrence of a thermal error may be connected to the device, for example, by a network. It is also possible to notify the management device and enable the remote management to grasp the status.
• the change of the operation content of the software and the operation adjustment of the hardware are not limited to the change method described above.
• control method list or reaction table 118 may not be fixed. For example, if a thermal error occurs frequently while the game application is operating, the operation will be made stronger, such as using a two-step control method simultaneously when the thermal error occurs while the game application is operating. You may change the control method list to suppress.
• control targets are the main processor 10 and the graphic processor 20, and the force that causes their internal temperature to be in the heat generation state, for example, processor surface, other places on the system board 100. Even if temperature sensors are provided, the average temperature of each temperature sensor may be included in the heat generation condition.
• the present invention is applicable to electronic devices such as computers, mobile phones, and game machines.

Landscapes

• Engineering & Computer Science (AREA)
• Theoretical Computer Science (AREA)
• Human Computer Interaction (AREA)
• Physics & Mathematics (AREA)
• General Engineering & Computer Science (AREA)
• General Physics & Mathematics (AREA)
• Debugging And Monitoring (AREA)
• Stored Programmes (AREA)
• Executing Machine-Instructions (AREA)

Priority Applications (1)

Applications Claiming Priority (2)

Publications (1)

Family

ID=37942493

Family Applications (1)

Country Status (3)

Cited By (1)

Families Citing this family (15)

Citations (2)

Family Cites Families (20)

• 2005
  • 2005-10-14 JP JP2005300507A patent/JP2007109085A/ja active Pending
• 2006
  • 2006-07-20 US US12/067,315 patent/US8068940B2/en active Active
  • 2006-07-20 WO PCT/JP2006/314385 patent/WO2007043231A1/ja not_active Ceased

Patent Citations (2)

Also Published As

Similar Documents

Legal Events