TY - GEN
T1 - SPM management using markov chain based data access prediction
AU - Yemliha, Taylan
AU - Srikantaiah, Shekhar
AU - Kandemir, Mahmut
AU - Ozturk, Ozean
PY - 2008
Y1 - 2008
N2 - Leveraging the power of scratchpad memories (SPMs) available in most embedded systems today is crucial to extract maximum performance from application programs. While regular accesses like scalar values and array expressions with affine subscript functions have been tractable for compiler analysis (to be prefetched into SPM), irregular accesses like pointer accesses and indexed array accesses have not been easily amenable for compiler analysis. This paper presents an SPM management technique using Markov chain based data access prediction for such irregular accesses. Our approach takes advantage of inherent, but hidden reuse in data accesses made by irregular references. We have implemented our proposed approach using an optimizing compiler. In this paper, we also present a thorough comparison of our different dynamic prediction schemes with other SPM management schemes. SPM management using our approaches produces 12.7% to 28.5% improvements in performance across a range of applications with both regular and irregular access patterns, with an average improvement of 20.8%.
AB - Leveraging the power of scratchpad memories (SPMs) available in most embedded systems today is crucial to extract maximum performance from application programs. While regular accesses like scalar values and array expressions with affine subscript functions have been tractable for compiler analysis (to be prefetched into SPM), irregular accesses like pointer accesses and indexed array accesses have not been easily amenable for compiler analysis. This paper presents an SPM management technique using Markov chain based data access prediction for such irregular accesses. Our approach takes advantage of inherent, but hidden reuse in data accesses made by irregular references. We have implemented our proposed approach using an optimizing compiler. In this paper, we also present a thorough comparison of our different dynamic prediction schemes with other SPM management schemes. SPM management using our approaches produces 12.7% to 28.5% improvements in performance across a range of applications with both regular and irregular access patterns, with an average improvement of 20.8%.
UR - http://www.scopus.com/inward/record.url?scp=57849143765&partnerID=8YFLogxK
UR - http://www.scopus.com/inward/citedby.url?scp=57849143765&partnerID=8YFLogxK
U2 - 10.1109/ICCAD.2008.4681632
DO - 10.1109/ICCAD.2008.4681632
M3 - Conference contribution
AN - SCOPUS:57849143765
SN - 9781424428205
T3 - IEEE/ACM International Conference on Computer-Aided Design, Digest of Technical Papers, ICCAD
SP - 565
EP - 569
BT - 2008 IEEE/ACM International Conference on Computer-Aided Design Digest of Technical Papers, ICCAD 2008
PB - Institute of Electrical and Electronics Engineers Inc.
T2 - 2008 International Conference on Computer-Aided Design, ICCAD
Y2 - 10 November 2008 through 13 November 2008
ER -