event

Ph.D. Dissertation Defense - Suk Chan Kang

Primary tabs

TitleOptimizing High Locality Memory References in Cache Coherent Shared Memory Multi-core Processors

Committee:

Dr. Sudhakar Yalamanchili, ECE, Chair , Advisor

Dr. Ada Gavrilovska, CoC

Dr. Linda Wills, ECE

Dr. Tushar Krishna, ECE

Dr. Santosh Pande, CS

Abstract:

The objective of this thesis research is to identify and optimize two classes of high locality data memory reference streams in cache coherent shared memory multi-processors. More specifically, this thesis classifies such memory objects into spatial and temporal false shared memory objects: modern cache coherent shared memory multi-processor systems make the assumption that every memory reference is a shared memory operation and, consequently, unconditionally prepare to incur the shared-memory-related overheads for every reference. This thesis explores two different schemes to minimize the shared memory abstraction overheads associated with these two target memory reference streams. The schemes implement the exception rules which enable isolating false memory objects from the shared memory domain, in a spatial and temporal manner. However, the exception rules definitely require special consideration in cache coherent shared memory multi-processors, regarding the data consistency, cache coherence, and memory consistency model. Thus, this thesis not only implements the schemes based on such consideration, but also breaks the chain of the widespread faulty assumption of prior academic work. This high-level approach ultimately aims at upgrading scalability of large scale systems, such as multi-socket cache coherent shared memory multi-processors, throughout improving performance and reducing energy/power consumption.

Status

  • Workflow Status:Published
  • Created By:Daniela Staiculescu
  • Created:12/14/2018
  • Modified By:Daniela Staiculescu
  • Modified:12/14/2018

Categories

Target Audience