Method and apparatus for autonomic policy-based thermal management in a data processing system

• US 7,194,645 B2
• Filed: 02/09/2005
• Issued: 03/20/2007
• Est. Priority Date: 02/09/2005
• Status: Expired due to Fees

• Current Assignee: International Business Machines Corporation
• Original Assignee: International Business Machines Corporation
• Inventors: Fields, James Stephen Jr., Nayar, Naresh, Eisen, Lee Evan, Floyd, Michael Stephen, Bieswanger, Andreas, McCredie, Bradley David
• Primary Examiner(s): BARNES-BULLOCK, CRYSTAL JOY
• Application Number:
  US11/054,261
• Publication Number:
  US 20060178764A1
• Time in Patent Office: 769 Days
• Field of Search: 700/11, 700/12, 700/21, 700/28, 700 32- 35, 700/54, 700/79, 700/81, 700/299, 702/130, 702/132, 702/136, 713/300, 713/320, 713322-324, 713/330, 713/340, 714/1, 714/2, 714/47, 714/23, 714/24, 340/500, 340/584, 361/687, 361/688
• US Class Current: 713/320
• CPC Class Codes: G06F 1/206 comprising thermal management

Abstract

A method, apparatus and computer instructions are provided to autonomically monitor and adjust system characteristics based on a customer optimization goal specified in a policy or profile. An autonomic management component is implemented in firmware comprising a set of control algorithms. Response to reading system characteristics from a plurality of sensors, the autononmic management component selects at least one control algorithm from the set and the control algorithm adjusts the parameters of the system characteristic to optimize performance according to the optimization goal specified by the customer.

Claims

1. A method for autonomic policy-based thermal management in a data processing system having at least one processor, the method comprising:

(1) selecting at least one algorithm from a plurality of algorithms to manipulate one or more effectors, in the data processing system, based on a user policy;

(2) in response to selecting an algorithm for performance, performing the following steps;

(3) monitoring a temperature of the data processing system to form a monitored temperature;

(4) determining if the monitored temperature is above a warning temperature;

(5) in response to the monitored temperature being above the warning temperature;

(a) determining if at least one effector is above a minimum acceptable value;

(b) in response to the at least one effector being above the minimum acceptable value, decreasing the at least one effector by an incremental amount according to a selected algorithm;

(c) in response to determining that the monitored temperature is above the warning temperature, returning to step (5)(a); and

(d) in response to determining that the monitored temperature is below the warning temperature, returning to step (3); and

(6) in response to the monitored temperature being below the warning temperature;

(e) determining if the at least one effector is below a maximum acceptable value;

(f) in response to the at least one effector being below the maximum acceptable value, increasing the at least one effector by the incremental amount according to the selected algorithm; and

(g) returning to step (4).

2. The method of claim 1 further comprising:

(7) in response to selecting an algorithm for acoustics, performing the following steps;

(8) monitoring the temperature of the data processing system to form the monitored temperature;

(9) determining if the monitored temperature is above the warning temperature;

(10) in response to the monitored temperature being above the warning temperature;

(h) determining if the at least one effector is below the maximum acceptable value;

(i) in response to the at least one effector being below the maximum acceptable value, increasing the at least one effector by the incremental amount according to the selected algorithm;

(j) in response to determining that the monitored temperature is above the warning temperature, returning to step (10)(h); and

(k) in response to determining that the monitored temperature is below the warning temperature, returning to step (8); and

(11) in response to the monitored temperature being below the warning temperature;

(1) determining if the at least one effector is above the minimum acceptable value;

(m) in response to the at least one effector being above the minimum acceptable value, decreasing the at least one effector by the incremental amount according to the selected algorithm; and

(n) returning to step (9).

3. The method of claim 2 further comprising:

(12) selecting an algorithm for optimizing both performance and acoustics; and

(13) adjusting the at least one effector to find an optimal tradeoff between the performance and the acoustics.

4. The method of claim 2, wherein the at least one effector comprises a voltage level, a frequency of the at least one processor, a fan speed, or a processor throttling.

5. The method of claim 1 further comprising:

(15) determining the incremental amount using one of a programmed and an empirically learned amount; and

(16) in response to adjusting the at least one effector by the incremental amount, waiting for a specified period of time.

6. The method of claim 1 further comprising:

(17) notifying the user if the monitored temperature is above the warning temperature and if one of the following conditions exists;

(i) at least one effector reaches the minimum acceptable operating level;

or

(ii) at least one effector reaches the maximum acceptable operating level.

7. The method of claim 1 further comprising:

(18) reading a value of the at least one effector from at least one sensor from a plurality of sensors in the data processing system, wherein the at least one sensor comprises an air-inlet temperature sensor, a processor temperature sensor, a power supply current sensor, or a time of day sensor.

8. The method of claim 1, wherein an automatic management component comprises the plurality of algorithms, and wherein the automatic management component performs the autonomic policy-based thermal management in the data processing system.

9. The method of claim 1, wherein the user policy dynamically changes.

10. A data processing system comprising:

at least one processor;

an autonomic management component coupled to a firmware in the data processing system, wherein the at least one processor executes the autonomic management component; and

at least one algorithm from a plurality of algorithms for manipulating one or more effectors, wherein the autonomic management component comprises the plurality of algorithms, and wherein the autonomic management component selects the at least one algorithm from the plurality of algorithms, and wherein the plurality of algorithms comprises an algorithm for performance which performs the following steps;

(1) monitoring a temperature of the data processing system to form a monitored temperature;

(2) determining if the monitored temperature is above a warning temperature;

(3) in response to the monitored temperature being above the warning temperature;

(a) determining if at least one effector is above a minimum acceptable value;

(b) in response to the at least one effector being above the minimum acceptable value, decreasing the at least one effector by an incremental amount according to the selected algorithm;

(c) in response to determining that the monitored temperature is above the warning temperature, returning to step (3)(a); and

(d) in response to determining that the monitored temperature is below the warning temperature, returning to step (1); and

(4) in response to the monitored temperature being below the warning temperature;

(e) determining if the at least one effector is below a maximum acceptable value;

(f) in response to the at least one effector being below the maximum acceptable value, increasing the at least one effector by the incremental amount according to the selected algorithm; and

(g) returning to step (2).

11. The data processing system of claim 10, wherein the plurality of control algorithms further comprises an algorithm for acoustics which performs the following steps:

(5) monitoring the temperature of the data processing system to form the monitored temperature;

(6) determining if the monitored temperature is above the warning temperature;

(7) in response to the monitored temperature being above the warning temperature;

(h) determining if the at least one effector is below the maximum acceptable value;

(i) in response to the at least one effector being below the maximum acceptable value, increasing the at least one effector by the incremental amount according to the selected algorithm;

(j) in response to determining that the monitored temperature is above the warning temperature, returning to step (7)(h); and

(k) in response to determining that the monitored temperature is below the warning temperature, returning to step (5); and

(8) in response to the monitored temperature being below the warning temperature;

(l) determining if the at least one effector is above the minimum acceptable value;

(m) in response to the at least one effector being above the minimum acceptable value, decreasing the at least one effector by the incremental amount according to the selected algorithm; and

(n) returning to step (6).

12. The data processing system of claim 11, wherein the plurality of algorithms further comprises an algorithm for optimizing both performance and acoustics, and wherein the algorithm for optimizing both performance and acoustics performs the step of adjusting the at least one effector to find an optimal tradeoff between the performance and the acoustics.

13. The data processing system of claim 11, wherein the at least one effector comprises a voltage level, a frequency of the at least one processor, a fan speed, or a processor throttling.

14. The data processing system of claim 10, wherein the at least one algorithm notifies the user if the monitored temperature level is above the warning temperature and if one of the following conditions exists:

(i) at least one effector reaches the minimum acceptable operating level;

or

(ii) at least one effector reaches the maximum acceptable operating level.

15. The data processing system of claim 10, wherein the at least one algorithm further performs the following steps:

(9) determining the incremental amount using one of a programmed and an empirically learned amount; and

(10) in response to adjusting the at least one effector by the incremental amount, waiting for a specific period of time.

16. A computer program product comprising a computer usable medium including instructions for autonomic policy-based thermal management in a data processing system having at least one processor, the computer program product comprising:

(1) instructions for selecting at least one algorithm from a plurality of algorithms to manipulate one or more effectors, in the data processing system, based on a user policy;

(2) in response to selecting an algorithm for performance, instructions for performing the following steps;

(3) instructions for monitoring a temperature of the data processing system to form a monitored temperature;

(4) instructions for determining if the monitored temperature is above a warning temperature;

(5) in response to the monitored temperature being above the warning temperature;

(a) instructions for determining if at least one effector is above a minimum acceptable value;

(b) in response to the at least one effector being above the minimum acceptable value, instructions for decreasing the at least one effector by an incremental amount according to the selected algorithm;

(c) in response to determining that the monitored temperature is above the warning temperature, instructions for returning to step (5)(a); and

(d) in response to determining that the monitored temperature is below the warning temperature, instructions for returning to step (3); and

(6) in response to the monitored temperature being below the warning temperature;

(e) instructions for determining if the at least one effector is below a maximum acceptable value;

(f) in response to the at least one effector being below the maximum acceptable value, instructions for increasing the at least one effector by an incremental amount according to the selected algorithm; and

(g) instructions for returning to step (4).

17. The computer program product of claim 16 further comprising:

(7) in response to selecting an algorithm for acoustics, instructions for performing the following steps;

(8) instructions for monitoring the temperature of the data processing system to form the monitored temperature;

(9) instructions for determining if the monitored temperature is above the warning temperature;

(10) in response to the monitored temperature being above the warning temperature;

(h) instructions for determining if the at least one effector is below the maximum acceptable value;

(i) in response to the at least one effector being below the maximum acceptable value, instructions for increasing the at least one effector by the incremental amount according to the selected algorithm;

(j) in response to determining that the monitored temperature is above the warning temperature, instructions for returning to step (10)(h); and

(k) in response to determining that the monitored temperature is below the warning temperature, instructions for returning to step (8); and

(11) in response to the monitored temperature being below the warning temperature;

(l) instructions for determining if the at least one effector is above the minimum acceptable value;

(m) in response to the at least one effector being above the minimum acceptable value, instructions for decreasing the at least one effector by an incremental amount according to the selected algorithm; and

(n) instructions for returning to step (9).

18. The computer program product of claim 17, wherein the at least one effector comprises a voltage level, a frequency of the at least one processor, a fan speed, or a processor throttling.

19. The computer program product of claim 16 further comprising:

(12) instructions for notifying the user if the monitored temperature level is above the warning temperature and if one of the following conditions exists;

(i) at least one effector reaches the minimum acceptable operating level;

or

(ii) at least one effector reaches the maximum acceptable operating level.

20. The computer program product of claim 16 further comprising:

(13) instructions for determining the incremental amount using one of a programmed and an empirically learned amount; and

(14) in response to adjusting the at least one effector by the incremental amount, instructions for waiting for a specific period of time.