Back to results
US 2010/0138609 A1Published application
Technique for controlling computing resources
AssigneeIntel Corporation
InventorsWilliam Hasenplaugh, Joel Emer, Tryggve Fossum, Aamer Jaleel, Simon Steely
Life of this patent
Estimating the term…
Abstract
A technique to enable resource allocation optimization within a computer system. In one embodiment, a gradient partition algorithm (GPA) module is used to continually measure performance and adjust allocation to shared resources among a plurality of data classes in order to achieve optimal performance.
Claims24
An apparatus comprising:
- a gradient partition algorithm (GPA) module to continually control allocation of at least one processing resource to at least one instruction thread, wherein the GPA module is to continually perform tests on the processing resource to determine a dynamic throttle setting that optimally meets the processing needs of the at least one instruction thread.
A system comprising:
- a processor core;
- a cache coupled to processor core;
- a gradient partition algorithm (GPA) module to continually perform tests on the cache to determine a dynamic throttle setting that optimally meets the processing needs of a plurality of instructions to access the cache.
A method comprising:
- partitioning a cache in at least two parts, in which to perform resource allocation tests;
- partitioning each of the at least two parts into a plurality of blocks corresponding to a plurality of data classes;
- continually performing the tests on each of the at least two parts and comparing their performance results, wherein the test is performed on one of the at least two parts at a first throttle setting and the test is performed on a second of the least two parts at a second throttle setting;
- adjusting the size of at least one of the plurality of blocks in response to the comparison of the performance results.
1 more element
Dependencies are inferred from the claim text.