Power Control of Servers Using Advanced Configuration and Power Interface (ACPI) States

• US 20090055665A1
• Filed: 08/22/2007
• Published: 02/26/2009
• Est. Priority Date: 08/22/2007
• Status: Active Grant

• Current Assignee: International Business Machines Corporation
• Original Assignee: International Business Machines Corporation
• Inventors: Maglione, Stephen C., Suffern, Edward Stanley
• Granted Patent: US 8,250,382 B2
• Time in Patent Office: Days
• US Class Current: 713/320
• CPC Class Codes: G06F 1/206 comprising thermal management
• CPC Class Codes: G06F 1/3203 Power management, i.e. event-based initiation of a power-saving mode
• CPC Class Codes: H05K 7/1498 Resource management, Optimisation arrangements, e.g. configuration, identification, tracking, physical location thermal management H05K7/20836
• CPC Class Codes: Y02D 10/00 Energy efficient computing, e.g. low power processors, power management or thermal management

Abstract

A method of managing power consumption by a plurality of blade servers within a processing system. The speed of at least one of the plurality of blade servers is reduced in response to the processing system reaching a power or thermal threshold. At least one of the plurality of blade servers is identified as not being critical to maintain in a working state and the critical blade server is put in a sleep state. A satellite management controller may control blade server power consumption and heat generation in various ways that combine processor speed-stepping and control of processor sleep states. Known sleep states save more power than speed-stepping by turning off the processor and/or volatile memory. The processor speed and sleep-states of at least one non-critical blade server, and optionally the processor speed of a critical processor, may be changed in order to control the power consumption below a power threshold or control the temperature below a thermal threshold.

Claims

1. A method of managing power consumed by a plurality of blade servers within a processing system, comprising:

identifying at least one of the plurality of blade servers that is not critical to maintain in a working state; and

putting a processor of the at least one non-critical blade server in a sleep state in response to determining that the processing system has reached a thermal or power threshold.

2. The method of claim 1, further comprising:

reducing the speed of the processor of at least one of the plurality of blade servers in response to determining that the processing system has reached a thermal or power threshold.

3. The method of claim 1, wherein the step of putting the processor of at least one non-critical blade server in a sleep state includes sending an IPMI message from a satellite management controller to command the BMC to put the processor of the at least one non-critical blade server in a sleep state.

4. The method of claim 1, wherein the sleep state is a standby state that maintains power to main memory within the device while power is shut off to a majority of other components within the device.

5. The method of claim 1, wherein the sleep state is a hibernation state that saves the content of main memory within the device to a hard drive and powers off the main memory.

6. The method of claim 3, wherein the step of putting the processor of the at least one non-critical blade server in a sleep state further includes sending an IPMI message from the BMC to command the operating system to put the processor of the at least one non-critical blade server in a sleep state.

7. The method of claim 2, further comprising:

changing the speed of the processor of the at least one non-critical blade server and changing the sleep state of at least one other non-critical blade server to control the power consumption below the power threshold or control the temperature below the thermal threshold.

8. The method of claim 7, further comprising:

putting all of the non-critical blade servers in a sleep state; and

changing the speed of at least one critical blade server to control the power consumption below the power threshold or control the temperature below the thermal threshold in response to determining that the step of putting the all of the non-critical blade servers in a sleep state was insufficient to control the power consumption below the power threshold or control the temperature below the thermal threshold.

9. The method of claim 8, further comprising:

turning off any or all of the non-critical servers to preserve available power for the critical servers remain operational.

10. A computer program product comprising a computer usable medium including computer usable program code for managing power to a plurality of blade servers within a processing system, the computer program product including:

computer usable program code for identifying at least one of the plurality of blade servers that is not critical to maintain in a working state; and

computer usable program code for putting the at least one non-critical blade server in a sleep state in response to determining that the processing system has reached a thermal or power threshold.

11. The computer program product of claim 10, further comprising:

computer usable program code for reducing the speed of at least one of the plurality of blade servers in response to determining that the processing system has reached a thermal or power threshold.

12. The computer program product of claim 10, wherein the computer usable program code for putting the at least one non-critical blade server in a sleep state includes computer usable program code for sending an IPMI message from a satellite management controller to command the BMC to put the at least one non-critical blade server in a sleep state.

13. The computer program product of claim 10, wherein the sleep state is a standby state that maintains power to main memory within the device while power is shut off to a majority of other components within the device.

14. The computer program product of claim 10, wherein the sleep state is a hibernation state that saves the content of main memory within the device to a hard drive and powers off the main memory.

15. The computer program product of claim 11, further comprising:

computer usable program code for changing the speed of at least one non-critical blade server and changing the sleep state of at least one other non-critical blade server to control the power consumption below the power threshold or control the temperature below the thermal threshold.

16. The computer program product of claim 15, further comprising:

computer usable program code for putting the all of the non-critical blade servers in a sleep state; and

computer usable program code for changing the speed of at least one critical blade server to control the power consumption below the power threshold or control the temperature below the thermal threshold in response to determining that the step of putting the all of the non-critical blade servers in a sleep state was insufficient to control the power consumption below the power threshold or control the temperature below the thermal threshold.

17. A power-controlled processing system, comprising:

a plurality of rack-mounted, ACPI-capable blade servers, each blade server having a processor running an operating system and a baseboard management controller in digital communication with the operating system;

a shared power supply providing electrical power to each of the plurality of blade servers;

a satellite management controller in digital communication with each of the baseboard management controllers, wherein the satellite management controller includes computer usable program code for identifying at least one of the plurality of processors that is not critical to maintain in a working state, and putting the at least one non-critical blade server in a sleep state in response to determining that the processing system has reached a thermal or power threshold.

18. The power-controlled processing system of claim 17, wherein the satellite management controller further includes computer usable program code for reducing the speed of at least one of the plurality of blade servers in response to determining that the processing system has reached a thermal or power threshold.

19. The power-controlled processing system of claim 17, wherein the computer usable program code for putting the at least one non-critical blade server in a sleep state includes computer usable program code for sending an IPMI message from a satellite management controller to command the BMC to put the at least one non-critical blade server in a sleep state.

20. The power-controlled processing system of claim 17, wherein the satellite management controller further includes computer usable program code for turning off any or all of the plurality of blade servers.