TWI700642B - Safety temperature control method and server - Google Patents

Abstract

A safe temperature control method is performed by a server, the server includes a power supply module and a motherboard, the motherboard includes a baseboard management controller and a system of chip, and the baseboard management controller setting a first critical temperature and a second critical temperature according to a specific temperature value, when the baseboard management controller determines that a current temperature is greater than the second critical temperature, the baseboard management controller controls the system of chip to operate in a underclocking mode, when determining the current temperature less than the first critical temperature, and the current temperature is between the first critical temperature and the second critical temperature, and the duration between the first critical temperature and the second critical temperature exceeds a predetermined time, the baseboard management controller controls the system of chip to operate in a power-off mode.

Description

Safe temperature control method and server

The present invention relates to a method and server for processing electronic digital data, in particular to a method and device for controlling the temperature of power supply.

When the existing server (ie, computer device) is operating, its power supply module will perform temperature monitoring when powering the system chip on the motherboard. At this time, the baseboard management controller (BMC: Baseboard Management Controller) can automatically The power management module obtains temperature monitoring related data, but does not perform any temperature control related processing, so it has the following disadvantages: Disadvantages 1. When the power management module detects that its own temperature exceeds the preset temperature, the power management module The group will shut itself down, so that the system chip will be shut down when there is no power supply. This will cause the entire system chip of the motherboard to restart without warning during operation and cause damage. Disadvantage 2: The data running before the restart will also cause data loss due to no prior backup. Therefore, how to improve the temperature control mechanism of the existing server is the current research direction.

Therefore, an object of the present invention is to provide a method that analyzes and compares the operating temperature of the power supply-related devices with the standard temperature during the operation of the system, and reduces the operating performance of the system or performs data backup and restarts the system as appropriate. Safe temperature control method.

Therefore, the safe temperature control method of the present invention is executed by a server, the server includes a power supply module for providing power, and a motherboard electrically connected to the power supply module to receive power, the power supply module Preset a standard temperature value for power supply, the motherboard includes a baseboard management controller that reads the standard temperature values, and a baseboard management controller that is electrically connected to the baseboard management controller and operates according to the power provided by the power supply module System chip. The substrate management controller respectively sets a first critical temperature not higher than the standard temperature value and a second critical temperature lower than the first critical temperature according to the standard temperature values. The system chip at least operates In one of a shutdown mode and a reduced load mode, the safe temperature control method is for the baseboard management controller to determine whether a current temperature is greater than the second critical temperature, and the current temperature is the current temperature of the power supply module If yes, the system chip operates in the load reduction mode according to the judgment result of the baseboard management controller, and the current temperature is the temperature corresponding to the power supply module when the power supply module supplies power to the system chip.

When the substrate management controller determines that the current temperature is less than the first critical temperature, the substrate management controller determines whether the current temperature is between the first critical temperature and the second critical temperature.

When the substrate management controller determines that the current temperature is between the first critical temperature and the second critical temperature, the substrate management controller determines that the current temperature is between the first critical temperature and the second critical temperature. Whether the time exceeds a predetermined time, when the judgment is yes, the system chip operates in the shutdown mode according to the judgment result of the baseboard management controller.

In addition, another object of the present invention is to provide a method for analyzing and comparing the operating temperature of the power supply-related devices with the standard temperature during the operation of the system, and reducing the operating performance of the system or performing data backup and restarting the system as appropriate. Server.

Therefore, the server of the present invention includes a power supply module and a motherboard.

The power supply module is used for supplying electric energy and presets a standard temperature value for power supply.

The motherboard is electrically connected to the power supply module, and includes a baseboard management controller that reads the standard temperature values, and a system chip that is electrically connected to the baseboard management controller and operates according to the power provided by the power supply module, The substrate management controller respectively sets a first critical temperature not higher than the standard temperature value and a second critical temperature lower than the first critical temperature according to the standard temperature value, and the system chip is operated in at least a shutdown mode One of two modes, and a load reduction mode.

The baseboard management controller determines whether a current temperature is greater than the second critical temperature, the current temperature is the current temperature of the power supply module, and if so, the system chip operates in the derating mode according to the determination result of the baseboard management controller , The current temperature is the temperature corresponding to the power supply module when the power supply module supplies power to the system chip.

When the substrate management controller determines that the current temperature is less than the first critical temperature, the substrate management controller determines whether the current temperature is between the first critical temperature and the second critical temperature.

When the substrate management controller determines that the current temperature is between the first critical temperature and the second critical temperature, the substrate management controller determines that the current temperature is between the first critical temperature and the second critical temperature. Whether the time exceeds a predetermined time, when the judgment is yes, the system chip operates in the shutdown mode according to the judgment result of the baseboard management controller.

The effect of the present invention is that the baseboard management controller sets the first and second critical temperature values according to the standard temperature value when the power supply module is powered, and when the power supply module actually supplies power, and the power supply mode When the temperature generated by the corresponding group exceeds the second critical temperature, the system chip operates in the load reduction mode. When the corresponding temperature generated by the power supply module is between the first and second critical temperatures, and the maintenance time exceeds For the predetermined time, the system chip is operated in the shutdown mode to prevent the system chip from being shut down directly and causing component damage and data loss in operation.

Referring to FIG. 1, the server of the present invention includes a power supply module 2, a motherboard 3, and a fan module 4.

The power supply module 2 is used to provide power to the motherboard 3 and preset a standard temperature value for power supply.

The motherboard 3 is electrically connected to the power supply module 2, and includes a baseboard management controller 31 that reads the standard temperature values, and a baseboard management controller 31 that is electrically connected to the baseboard management controller 31 and is based on the power provided by the power supply module 2. The operating system chip 32, and a peripheral hardware 33 that is electrically connected to the baseboard management controller 31 and controlled by the baseboard management controller 31 to operate. The baseboard management controller 31 sets a different temperature value according to the standard temperature value. A first critical temperature higher than the standard temperature value, a second critical temperature lower than the first critical temperature, and a third critical temperature lower than the second critical temperature. The system chip 32 selectively operates in one of a shutdown mode, a load reduction mode, and a non-load reduction mode. The system chip 32 further includes a platform path control electrically connected to the baseboard management controller 31 PCH (Platform Controller Hub) 321, and a central processing unit 322 that is electrically connected to the platform path controller 311 through a direct media interface (DMI: Direct Media Interface) (ie, a bus).

It should be added that, in this embodiment, the system on chip (SoC: System on Chip) 32 is the platform path controller 321 and the CPU 322 described above, but it is not limited to this embodiment. In the same way, for example, the system chip 32 can also be operated solely by the central processing unit. In addition, the baseboard management controller 31 of this embodiment is in the form of firmware, and the baseboard management controller 31 executes its firmware program to operate.

The fan module is electrically connected to the platform path controller 321 and the peripheral hardware 33, and operates by receiving the power provided by the power supply module 2 to the motherboard to dissipate heat from the entire motherboard related components.

Referring to FIG. 2 in conjunction with FIG. 3, a safe temperature control method executed by the server is further described below. The safe temperature control method includes a step S2 of reading the actual temperature, a step S3 of judging the temperature of the first and second intervals, A high temperature derating step S4, a first critical temperature determination step S5, a system event recording step S6, a shutdown step S7, a second interval temperature determination step S8, a temperature maintenance time step S9, a first determination step And the third interval temperature step S10, a release low temperature derating step S11, an executing system derating step S12, and a second and fourth interval temperature determination step S13.

It should be noted that, FIG. 3 shows the possible distribution intervals of the corresponding temperature generated when the power supply module 2 supplies power, and the operation mode of the CPU 322 correspondingly switched in each interval.

The actual temperature reading step S2 is that when the power supply module 2 supplies power to the motherboard, the baseboard management controller 31 reads a current temperature of the power supply module 2, and the current temperature is the power supply module When power is supplied to the system chip 32, the power supply module 2 corresponds to the temperature generated. In addition, in this embodiment, the machine version management controller 3 continuously reads and determines the current temperature of the power supply module 2 in a specific period. The temperature is used by the central processing unit to make corresponding mode switching, which is described in further detail below.

The step S3 of determining the first and second interval temperatures is for the baseboard management controller 31 to determine whether the current temperature is greater than the second critical temperature, and if the determination result is no, return to the step S2 of reading the actual temperature.

The step S4 of performing high temperature derating is that when the baseboard management controller 31 determines that the current temperature is greater than the second critical temperature, the system chip 32 operates in the derating mode according to the determination result of the baseboard management controller 31, and in more detail In other words, if the current temperature is greater than the second critical temperature, it means that the current temperature may fall in the first zone or the second zone (FIG. 3), the system chip 32 is operating in the reduced load mode, and the CPU 322 is reduced in operation. Efficiency to reduce the power supply of the power supply module 2 (for example: frequency reduction), that is, the baseboard management controller 31 sends a relative notification signal to the platform path controller 321, and then triggers the ME firmware of the platform path controller 321 The body executes the corresponding program, thereby causing the central processing unit 322 to perform load reduction. That is, the baseboard management controller 31 is specifically a firmware-based architecture. When it determines that the temperature has changed, it first sends out a related notification signal to the central processing unit 322 via the platform path controller 321 Therefore, the central processing unit 322 switches the operation mode. When the temperature changes later, the operating mechanism between the baseboard management controller 31 and the central processing unit 322 is the same, so it will not be repeated.

Then the step S5 of determining the first critical temperature is that the BMC 31 then continues to periodically read the current temperature, and determines whether the current temperature exceeds the first critical temperature (after frequency reduction, monitor the power supply module Whether the temperature of the power supply drops).

The recording system event step S6 is when the baseboard management controller 31 determines that the current temperature exceeds the first critical temperature, that is, when the power supply module 2 reduces the power supply, but the temperature does not drop, then the baseboard management controller 31 31 Record the relevant system event log (SEL: System Event Log) first.

In the shutdown step S7, the system chip 32 is then operated in the shutdown mode according to the judgment result of the baseboard management controller 31, so that the entire system enters the shutdown state. Furthermore, as explained in the step S4 of performing high temperature reduction, The baseboard management controller 31 sends a relative notification signal to the platform path controller 321, and then triggers the management engine (ME: Management Engine) firmware of the platform path controller 321 to execute the corresponding program, thereby enabling the central processing unit 322 Perform shutdown.

The step S8 of determining the temperature in the second interval is that when the baseboard management controller 31 determines that the current temperature does not exceed the first critical temperature (not falling in the first interval of FIG. 3), the baseboard management controller 31 then continues to periodically Read the current temperature, and determine whether the current temperature is between the first critical temperature and the second critical temperature (the second interval in FIG. 3).

The determining temperature maintaining time step S9 is when the baseboard management controller 31 determines that the current temperature is between the first critical temperature and the second critical temperature (that is, the current temperature falls within the second interval of FIG. 3), the The baseboard management controller determines whether the current temperature is maintained in the second interval for more than a predetermined time (for example, 10 minutes).

The step S10 of determining the temperature in the first and third intervals is that when the baseboard management controller 31 determines that the current temperature is not between the first critical temperature and the second critical temperature, the baseboard management controller 31 continues to periodically Read the current temperature, and determine whether the current temperature is between the second critical temperature and the third critical temperature (the third interval in FIG. 3), or whether the current temperature exceeds the first critical temperature.

The step S11 of releasing low-temperature load reduction is when the baseboard management controller 31 determines that the current temperature is not between the second critical temperature and the third critical temperature, and does not exceed the first critical temperature, then the system chip 32 According to the judgment result of the baseboard management controller 31, it operates in the non-load-down mode (representing that the temperature corresponding to the power supply module at this time falls within the fourth interval of FIG. 3), that is, the central processing unit 322 operates at normal efficiency Operation, there is no need to lower the clock, and then return to the step S2 of reading the actual temperature.

In the step S12 of executing the system down load, the baseboard management controller 31 determines that the current temperature is between the second critical temperature and the third critical temperature, and the system chip 32 operates in accordance with the determination result of the baseboard management controller 31 The load reduction mode.

The step S13 of determining the second and fourth interval temperature is that when the system chip 32 is controlled to operate in the reduced load mode according to the determination result of the baseboard management controller, the baseboard management controller 31 then continues to periodically read the current Temperature and determine whether the current temperature is between the first critical temperature and the second critical temperature, or whether the current temperature is lower than one of the third critical temperature, when the baseboard management controller 31 determines the current If the temperature is not between the first critical temperature and the second critical temperature, and is not lower than the third critical temperature, return to the execution system derating step S12, when the baseboard management controller 31 determines the current temperature Between the first critical temperature and the second critical temperature, return to the step S8 of determining the first interval temperature, and when the baseboard management controller 31 determines that the current temperature is lower than the third critical temperature, return to the The low-temperature derating step S11 is released.

In the above embodiment, the baseboard management controller 31 first reads the standard temperature value preset by the power supply module 2 itself, and accordingly presets three first to third critical temperatures that are not greater than the standard temperature value ( The three are arranged in sequence from high temperature to low temperature). When the power supply module 2 supplies power, the baseboard management controller 31 determines the current temperature of the power supply module 2 to perform different treatments, so that the system chip 32 is To operate in different modes, when the baseboard management controller 31 determines that the current temperature of the power supply module 2 is greater than the second critical temperature, the system chip 32 first runs under reduced load, and if the load is reduced, the power supply module 2’s If the temperature is greater than the first critical temperature, the treatment process is S2, S3, S4, S5, S6, S7 in order. If it is determined that the temperature of the power supply module 2 does not exceed the first critical temperature after the load is reduced, the substrate management The controller 31 then continues to periodically determine whether the temperature is between the first and second critical temperatures. If it is, and the time between the temperature interval exceeds the predetermined time, the treatment process is S2, S3, S4 in order. , S5, S8, S9, S6, if it is judged that the temperature is not between the first and second critical temperatures, then judge whether the temperature at this time is between the second to third critical temperatures, or whether it exceeds the first A critical temperature is one of the two. If it is not one of the two, the treatment process is S2, S3, S4, S5, S8, S10, S11 in order, if it is judged to be between the second and third critical temperatures The treatment process is S2, S3, S4, S5, S8, S10, S12 in order. If the judgment result is that the first critical temperature is exceeded, the treatment process is S2, S3, S4, S5, S8, S10, S6, S). In addition, when the baseboard management controller 31 determines that the time when the temperature of the power supply 2 is between the first and second critical temperatures does not exceed the predetermined time, the treatment process is S2, S3, S4, S5, S8, S9, S10. As for when the system chip 32 operates in the de-load mode, the baseboard management controller 31 then continues to periodically determine whether the current temperature of the power supply module 2 is within the first 1. Between the second critical temperature, or lower than the third critical temperature, if the judgment result is not one of the two, the treatment process will be S12, S13, S12 in order, if the judgment result is medium Between the first and second critical temperatures, the treatment process is S12, S13, and S8 in sequence. If the judgment result is lower than the third critical temperature, the treatment process is S12, S13, S11, and S2 in sequence.

To sum up, when the present invention executes the safety temperature control method, the baseboard management controller first reads the temperature specification reserved inside the power supply module, and sets the first threshold related to the temperature specification in advance according to the temperature specification The temperature and the second critical temperature. When the power supply module supplies power to the system chip, the baseboard management controller periodically monitors the current corresponding temperature of the power supply module. When the measured temperature exceeds the second critical temperature When temperature, the system chip first enters the load reduction mode, that is, the system chip is controlled to reduce the operating efficiency to reduce the power supply of the power supply module and reduce the temperature. After the load is reduced, the temperature measured by the baseboard management controller is greater than For the first critical temperature, the baseboard management controller controls the system chip to perform shutdown, that is, enters the shutdown mode, so that the user can decide whether to back up data in advance according to the overall system temperature change and prolong the service life of the overall related components. Make the overall system management mechanism more flexible, so it can indeed achieve the purpose of the invention.

However, the above are only examples of the present invention. When the scope of implementation of the present invention cannot be limited by this, all simple equivalent changes and modifications made in accordance with the scope of the patent application of the present invention and the content of the patent specification still belong to Within the scope of the patent for the present invention.

2: Power supply module

3: Motherboard

31: baseboard management controller

32: system chip

321: Platform Path Controller

322: Central Processing Unit

33: Peripheral hardware

4: Fan module

S2: Read the actual temperature step

S3: Judging the first critical temperature step

S4: Perform high temperature load reduction steps

S5: Judging the second critical temperature step

S6: Steps to record system events

S7: shutdown steps

S8: Steps to determine the temperature in the first interval

S9: Determine the temperature maintenance time step

S10: Steps to determine the second interval temperature

S11: Remove the load reduction step

S12: Perform low-temperature load reduction steps

S13: Steps to determine the temperature of the third interval

Other features and effects of the present invention will be clearly presented in the embodiments with reference to the drawings, in which: Figure 1 is a block diagram illustrating an embodiment of the server of the present invention; Figure 2 is a flowchart illustrating a safe temperature control method implemented by the invention; and FIG. 3 is a temperature distribution diagram to assist in explaining the temperature monitoring points and temperature distribution monitoring intervals related to the implementation of the safety temperature control method in this embodiment.

S2: Read the actual temperature step

S3: Judging the first critical temperature step

S4: Perform high temperature load reduction steps

S5: Judging the second critical temperature step

S6: Steps to record system events

S7: shutdown steps

S8: Steps to determine the temperature in the first interval

S9: Determine the temperature maintenance time step

S10: Steps to determine the second interval temperature

S11: Remove the load reduction step

S12: Perform low-temperature load reduction steps

S13: Steps to determine the temperature of the third interval

Claims (12)

A safe temperature control method is executed by a server. The server includes a power supply module for providing power, and a motherboard electrically connected to the power supply module to receive power. The power supply module is preset Regarding the standard temperature values during power supply, the motherboard includes a baseboard management controller that reads the standard temperature values, and a system that is electrically connected to the baseboard management controller and operates according to the power provided by the power supply module Chip, the substrate management controller respectively sets a first critical temperature not higher than the standard temperature value and a second critical temperature lower than the first critical temperature according to the standard temperature value, and the system chip operates at at least one One of a shutdown mode and a reduced load mode, the safe temperature control method includes: the baseboard management controller determines whether a current temperature is greater than the second critical temperature, and the current temperature is the current temperature of the power supply module; If yes, the system chip operates in the load reduction mode according to the judgment result of the baseboard management controller, and the current temperature is the temperature generated by the power supply module when the power supply module supplies power to the system chip; When the management controller determines that the current temperature is less than the first critical temperature, the substrate management controller determines whether the current temperature is between the first critical temperature and the second critical temperature; and when the substrate management controller determines the When the current temperature is between the first critical temperature and the second critical temperature, the substrate management controller determines whether the current temperature is between the first critical temperature and the second critical temperature for more than a predetermined time, when If the judgment is yes, the system chip operates in the shutdown mode according to the judgment result of the baseboard management controller. The safety temperature control method according to claim 1, wherein when the baseboard management controller determines that the current temperature is greater than the second critical temperature, the baseboard management controller continues to operate after the system chip operates in the load reduction mode accordingly Periodically read the current temperature and determine whether the current temperature exceeds the first critical temperature, so that the system chip determines the corresponding operation mode. According to the safety temperature control method of claim 2, when the baseboard management controller determines that the current temperature is between the first critical temperature and the second critical temperature for longer than the predetermined time, the system chip is operated during the shutdown Before the mode, the baseboard management controller also records a related event log. According to the safety temperature control method of claim 2, when the baseboard management controller determines that the current temperature is greater than the first critical temperature, the baseboard management controller records a related event log, and the system chip is controlled according to the baseboard management The judgment result of the device is operated in the shutdown mode. According to the safety temperature control method of claim 2, the baseboard management controller further sets a third critical temperature that is less than the second critical temperature according to the standard temperature value, and when the baseboard management controller determines that the current temperature is not between When between the first critical temperature and the second critical temperature, the substrate management controller determines whether the current temperature is between the second critical temperature and the third critical temperature, and greater than the first critical temperature In one of them, when it is determined that the current temperature is greater than the first critical temperature, the baseboard management controller records a related event log, and the system chip operates in the shutdown mode according to the determination result of the baseboard management controller. The safety temperature control method of claim 1, wherein when the substrate management controller determines that the current temperature is not greater than the second critical temperature, the substrate management The controller continues to periodically read the current temperature so that the system chip determines the corresponding operation mode. According to the safety temperature control method of claim 5, the system chip is also operated in a non-derating mode, wherein, when the baseboard management controller determines that the current temperature is not between the second critical temperature and the third critical temperature Between and not greater than the first critical temperature, the system chip operates in the non-load mode according to the judgment result of the baseboard management controller, and the baseboard management controller continues to periodically read the current temperature. According to the safety temperature control method of claim 5, when the baseboard management controller determines that the current temperature is between the second critical temperature and the third critical temperature, the system chip is based on the determination result of the baseboard management controller Operating in the derating mode, the baseboard management controller determines whether the current temperature is between the first critical temperature and the second critical temperature, and is lower than one of the third critical temperature, when the The baseboard management controller determines that the current temperature is not between the first critical temperature and the second critical temperature, and is not lower than the third critical temperature, and the system chip continues to operate in accordance with the determination result of the baseboard management controller The load reduction mode. According to the safety temperature control method of claim 7, the baseboard management controller determines that the current temperature is lower than the third critical temperature, the system chip operates in the non-derating mode according to the determination result of the baseboard management controller, and the substrate The management controller continues to periodically read the current temperature. According to the safety temperature control method of claim 8, the substrate management controller determines that the current temperature is between the first critical temperature and the second critical temperature, and the substrate management controller continues to periodically read the current temperature Temperature, and It is determined whether the current temperature is between the first critical temperature and the second critical temperature. According to the safety temperature control method of claim 10, if the judgment is yes, and the substrate management controller judges that the time between the first critical temperature and the second critical temperature does not exceed the predetermined time, the substrate The management controller continues to periodically determine whether the current temperature is between the second critical temperature and the third critical temperature, and is greater than one of the first critical temperature. A server includes: a power supply module for providing electric energy and preset a standard temperature value for power supply; and a motherboard electrically connected to the power supply module, including a reading of the standard temperature Value of the baseboard management controller, and a system chip that is electrically connected to the baseboard management controller and operates according to the power provided by the power supply module, and the baseboard management controller sets a temperature not higher than the standard temperature according to the standard temperature value Value of the first critical temperature and a second critical temperature lower than the first critical temperature, the system chip operates in at least one of a shutdown mode and a reduced load mode, and the baseboard management controller determines a current Whether the temperature is greater than the second critical temperature, the current temperature is the current temperature of the power supply module, if so, the system chip operates in the load reduction mode according to the judgment result of the baseboard management controller, and the current temperature is the power supply When the module supplies power to the system chip, the power supply module corresponds to the temperature generated, and when the baseboard management controller determines that the current temperature is less than the first temporary temperature When the boundary temperature is reached, the substrate management controller determines whether the current temperature is between the first critical temperature and the second critical temperature, and when the substrate management controller determines that the current temperature is between the first critical temperature and the first critical temperature When the second critical temperature is between, the substrate management controller determines whether the current temperature is between the first critical temperature and the second critical temperature for more than a predetermined time, and when the determination is yes, the system chip is controlled according to the substrate management The judgment result of the device operates in the shutdown mode.

Images