Thermal-And Spatial-Aware Task Scheduling

• US 20170277564A1
• Filed: 03/25/2016
• Published: 09/28/2017
• Est. Priority Date: 03/25/2016
• Status: Active Grant

• Current Assignee: International Business Machines Corporation
• Original Assignee: International Business Machines Corporation
• Inventors: Coteus, Paul W., Lee, Eun Kyung, Park, Yoonho
• Granted Patent: US 9,817,697 B2
• Time in Patent Office: Days
• CPC Class Codes: G06F 9/5094 where the allocation takes into account power or heat criteria power management in computers in general G06F1/3203; thermal management in computers in general G06F1/206
• CPC Class Codes: Y02D 10/00 Energy efficient computing, e.g. low power processors, power management or thermal management

Abstract

A method, apparatus, and computer program product are provided for thermal- and spatial-aware task scheduling. The method may include monitoring a temperature for each core of a central processing unit having a plurality of cores; determining, from the monitoring, a set of hotspot cores from the plurality of cores determining temperature information and distance information for each hotspot core in the set of hotspot cores relative to each of the other cores on the central processing unit; calculating a placement metric for each core of the central processing unit based at least on the determined distance information and the determined temperature information; and scheduling a task by allocating the task to one or more cores of the central processing unit according to the placement metric.

Claims

1. A method comprising:

monitoring a temperature for each core of a central processing unit having a plurality of cores;

determining, from the monitoring, a set of hotspot cores from the plurality of cores;

determining temperature information and distance information for each hotspot core in the set of hotspot cores relative to each of the other cores on the central processing unit;

calculating a placement metric for each core of the central processing unit based at least on the determined distance information and the determined temperature information; and

scheduling a task by allocating the task to one or more cores of the central processing unit according to the placement metric;

wherein the temperature information comprises a temperature vector, T, for each hotspot core in the set of hotspot cores and wherein the distance information comprises a distance vector, D, for each hotspot core in the set of hotspot cores, wherein calculating the placement metric is based on a platform dependent coefficient, c, indicative of heat conduction of the central processing unit, and wherein the number of hotspot cores in the set of hotspot cores, h, is user defined, and the placement metric is calculated by;

Σ

_i=1^h(c*D_iⁿ+T_i), where n is a number greater than zero whose value is related to a thermal conduction of at least the central processing unit.

3. The method of claim 1, wherein each temperature vector comprises temperature values indicative of a difference in temperature between a given hotspot core in the set of hotspot cores and each of the other cores of the central processing unit, and wherein each distance vector comprises distance values indicative of a distance between a given hotspot core in the set of hotspot cores and each of the other cores of the central processing unit.

6. The method of claim 1, wherein the platform dependent coefficient is at least one of:

a scaler value and a vector.

7. The method of claim 3, wherein each distance vector is predefined based on an architecture of the central processing unit.

8. The method of claim 1, wherein the temperature information is updated in response to arrival of a task to be scheduled.

9. The method according to claim 1, wherein each hotspot core in the set of hotspot cores comprises at least one of:

a core having the highest monitored temperature;

a core having a monitored temperature that is higher than adjacent cores; and

a core having a monitored temperature higher than a predefined threshold temperature.

2. (canceled)

4-5. -5. (canceled)

10. An apparatus comprising:

at least one processor; and

at least one non-transitory memory including computer program code;

the at least one memory and the computer program code may be configured to, with the at least one processor, cause the apparatus to;

monitor a temperature for each core of a central processing unit having a plurality of cores;

determine, from the monitoring, a set of hotspot cores from the plurality of cores;

determine temperature information and distance information for each respective hotspot core in the set of hotspot cores relative to each of the other cores on the central processing unit;

calculate a placement metric for each core of the central processing unit based at least on the determined distance information and the determined temperature information; and

schedule a task by allocating the task to one or more cores of the central processing unit according to the placement metric;

wherein the temperature information comprises a temperature vector, T, for each hotspot core in the set of hotspot cores and wherein the distance information comprises a distance vector, D, for each hotspot core in the set of hotspot cores, wherein calculation of the placement metric is based on a platform dependent coefficient, c, indicative of heat conduction of the central processing unit, and wherein the number of hotspot cores in the set of hotspot cores, h, is user defined, and the placement metric is calculated by;

Σ

_i=1^h(c*D_iⁿ+T_i), where n is a number greater than zero whose value is related to a thermal conduction of at least the central processing unit.

12. The apparatus of claim 10, wherein each temperature vector comprises temperature values indicative of a difference in temperature between a given hotspot core in the set of hotspot cores and each of the other cores of the central processing unit, and wherein each distance vector comprises distance values indicative of a distance between a given hotspot core in the set of hotspot cores and each of the other cores of the central processing unit.

15. The apparatus of claim 10, wherein the platform dependent coefficient is at least one of:

a scaler value and a vector.

16. The apparatus of claim 12, wherein each distance vector is predefined based on an architecture of the central processing unit.

17. The apparatus of claim 10, wherein the temperature information is updated in response to arrival of a task to be scheduled.

18. The apparatus according to claim 10, wherein each hotspot core in the set of hotspot cores comprises at least one of:

a core having the highest monitored temperature;

a core having a monitored temperature that is higher than adjacent cores; and

a core having a monitored temperature higher than a predefined threshold temperature.

11. (canceled)

13-14. -14. (canceled)

19. A computer program product for scheduling tasks, the computer program product comprising a computer readable storage medium having program instructions embodied therewith, the program instructions executable by a device to cause the device to:

monitor a temperature for each core of a central processing unit having a plurality of cores;

determine, from the monitoring, a set of hotspot cores from the plurality of cores;

determine temperature information and distance information for each respective hotspot core in the set of hotspot cores relative to each of the other cores on the central processing unit;

calculate a placement metric for each core of the central processing unit based at least on the determined distance information and the determined temperature information; and

schedule a task by allocating the task to one or more cores of the central processing unit according to the placement metric,

wherein the temperature information comprises a temperature vector, T, for each hotspot core in the set of hotspot cores and wherein the distance information comprises a distance vector, D, for each hotspot core in the set of hotspot cores, wherein calculation of the placement metric is based on a platform dependent coefficient, c, indicative of heat conduction of the central processing unit, and wherein the number of hotspot cores in the set of hotspot cores, h, is user defined, and the placement metric is calculated by;

Σ

_i=1^h(c*D_iⁿ+T_i), where n is a number greater than zero whose value is related to a thermal conduction of at least the central processing unit.

21. The computer program product of claim 19, wherein each temperature vector comprises temperature values indicative of a difference in temperature between a given hotspot core in the set of hotspot cores and each of the other cores of the central processing unit, and wherein each distance vector comprises distance values indicative of a distance between a given hotspot core in the set of hotspot cores and each of the other cores of the central processing unit.

22. The computer program product of claim 21, wherein each distance vector is predefined based on an architecture of the central processing unit.

23. The computer program product of claim 19, wherein the platform dependent coefficient is at least one of:

a scaler value and a vector.

24. The computer program product of claim 19, wherein the temperature information is updated in response to arrival of a task to be scheduled.

25. The computer program product of claim 19, wherein each hotspot core in the set of hotspot cores comprises at least one of:

a core having the highest monitored temperature;

a core having a monitored temperature that is higher than adjacent cores; and

a core having a monitored temperature higher than a predefined threshold temperature.

20. (canceled)