The Community for Technology Leaders
Green Image
Issue No. 02 - July-Dec. (2012 vol. 11)
ISSN: 1556-6056
pp: 33-36
Hyesoon Kim , Georgia Institute of Technology, Atlanta
Jinwoo Shin , Georgia Institute of Technology, Atlanta
Jaekyu Lee , Georgia Institute of Technology, Atlanta
Nagesh B. Lakshminarayana , Georgia Institute of Technology, Atlanta
ABSTRACT
GPGPU architectures (applications) have several different characteristics compared to traditional CPU architectures (applications): highly multithreaded architectures and SIMD-execution behavior are the two important characteristics of GPGPU computing. In this paper, we propose a potential function that models the DRAM behavior in GPGPU architectures and a DRAM scheduling policy α-SJF policy to minimize the potential function. The scheduling policy essentially chooses between SJF and FR-FCFS at run-time based on the number of requests from each thread and whether the thread has a row buffer hit.
INDEX TERMS
Random access memory, Processor scheduling, Computer architecture, Benchmark testing, Mathematical model, Equations, Instruction sets, DRAM scheduling, Random access memory, Processor scheduling, Computer architecture, Benchmark testing, Mathematical model, Equations, Instruction sets, Potential function, GPGPU
CITATION
Hyesoon Kim, Jinwoo Shin, Jaekyu Lee, Nagesh B. Lakshminarayana, "DRAM Scheduling Policy for GPGPU Architectures Based on a Potential Function", IEEE Computer Architecture Letters, vol. 11, no. , pp. 33-36, July-Dec. 2012, doi:10.1109/L-CA.2011.32
191 ms
(Ver )