Optimizing thermal performance using thermal flow analysis

• US 8,037,893 B2
• Filed: 05/03/2006
• Issued: 10/18/2011
• Est. Priority Date: 05/03/2006
• Status: Active Grant

• Current Assignee: Marvell Asia Pte Limited (Marvell Technology Group Limited)
• Original Assignee: International Business Machines Corporation
• Inventors: Johns, Charles Ray, Nutter, Mark Richard, Stafford, James Michael, Aguilar, Maximino Jr.
• Primary Examiner(s): LEE, CHUN KUAN
• Application Number:
  US11/381,373
• Publication Number:
  US 20070260415A1
• Time in Patent Office: 1,994 Days
• Field of Search: 137/12, 713/300
• US Class Current: 137/12
• CPC Class Codes: G06F 8/45 Exploiting coarse grain parallelism in compilation, i.e. parallelism between groups of instructions
• CPC Class Codes: G06F 9/4893 taking 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
• CPC Class Codes: Y10T 137/0379 By fluid pressure

Abstract

A computer implemented method and system for optimizing thermal performance of a computer system. An identification of a set of processor cores associated with the computer system is made and a thermal index is requested for each of the set of processor cores to form a set of thermal indexes. Proximity information and conductive property information associated with the set of processors is loaded and software is mapped to execute on an optimal processor core form the set of processor cores based the set of thermal indexes, proximity information, and conductive property information.

Claims

1. A computer implemented method for improving thermal performance of a computer system, the computer implemented method comprising:

identifying a set of processor cores associated with the computer system;

requesting a thermal index for each of the set of processor cores to form a set of thermal indexes, the thermal index representing a thermal efficiency for each of the processor cores in the set of processor cores and being generated by a compiler;

loading proximity information and conductive property information associated with the set of processors;

mapping software to execute on an optimal processor core from the set of processor cores based the set of thermal indexes, the proximity information, and the conductive property information to form mapped software;

storing the proximity information and the conductive property information in a data structure, wherein the proximity information defines the proximity of one processor core to another processor core in the set of processor cores and the conductive property information defines material used in manufacturing the processor cores; and

implementing corrective measures before executing the mapped software, wherein the corrective measures comprise modifying an instruction stream and reducing a bandwidth allocation to avoid hardware throttling.

2. A data processing system comprising:

a bus system;

a communications system connected to the bus system;

a memory connected to the bus system, wherein the memory includes an executable set of instructions; and

a processing unit connected to the bus system, wherein the processing unit executes the executable set of instructions, the set of instructions identifying a set of processor cores associated with the computer system;

requesting a thermal index for each of the set of processor cores to form a set of thermal indexes, the thermal index representing a thermal efficiency for each of the processor cores in the set of processor cores and being generated by a compiler;

loading proximity information and conductive property information associated with the set of processors, wherein the proximity information defines the proximity of one processor core to another processor core in the set of processor cores and the conductive property information defines material used in manufacturing the processor cores;

mapping software to execute on an optimal processor core from the set of processor cores based the set of thermal indexes, the proximity information, and the conductive property information to form mapped software;

storing the proximity information and the conductive property information in a data structure; and

implementing corrective measures before executing the mapped software, wherein the corrective measures comprise modifying an instruction stream and reducing a bandwidth allocation to avoid hardware throttling.

3. A computer program product for improving thermal performance of a computer system, the computer program product comprising:

a set of computer readable storage devices storing computer executable program code;

computer executable program code identifying a set of processor cores associated with the computer system;

computer executable program code requesting a thermal index for each of the set of processor cores to form a set of thermal indexes, the thermal index representing a thermal efficiency for each of the processor cores in the set of processor cores and being generated by a compiler;

computer executable program code loading proximity information and conductive property information associated with the set of processors;

computer executable program code mapping software to execute on an optimal processor core from the set of processor cores based the set of thermal indexes, the proximity information, and the conductive property information to form mapped software;

computer executable program code storing the proximity information and the conductive property information in a data structure, wherein the proximity information defines the proximity of one processor core to another processor core in the set of processor cores and the conductive property information defines material used in manufacturing the processor cores; and

computer executable program code implementing corrective measures before executing the mapped software, wherein the corrective measures comprise modifying an instruction stream and reducing a bandwidth allocation to avoid hardware throttling.