TY - GEN

T1 - Towards optimizing energy costs of algorithms for shared memory architectures

AU - Korthikanti, Vijay Anand

AU - Agha, Gul

PY - 2010

Y1 - 2010

N2 - Energy consumption by computer systems has emerged as an important concern. However, the energy consumed in executing an algorithm cannot be inferred from its performance alone: it must be modeled explicitly. This paper analyzes energy consumption of parallel algorithms executed on shared memory multicore processors. Specifically, we develop a methodology to evaluate how energy consumption of a given parallel algorithm changes as the number of cores and their frequency is varied. We use this analysis to establish the optimal number of cores to minimize the energy consumed by the execution of a parallel algorithm for a specific problem size while satisfying a given performance requirement. We study the sensitivity of our analysis to changes in parameters such as the ratio of the power consumed by a computation step versus the power consumed in accessing memory. The results show that the relation between the problem size and the optimal number of cores is relatively unaffected for a wide range of these parameters.

AB - Energy consumption by computer systems has emerged as an important concern. However, the energy consumed in executing an algorithm cannot be inferred from its performance alone: it must be modeled explicitly. This paper analyzes energy consumption of parallel algorithms executed on shared memory multicore processors. Specifically, we develop a methodology to evaluate how energy consumption of a given parallel algorithm changes as the number of cores and their frequency is varied. We use this analysis to establish the optimal number of cores to minimize the energy consumed by the execution of a parallel algorithm for a specific problem size while satisfying a given performance requirement. We study the sensitivity of our analysis to changes in parameters such as the ratio of the power consumed by a computation step versus the power consumed in accessing memory. The results show that the relation between the problem size and the optimal number of cores is relatively unaffected for a wide range of these parameters.

KW - Energy

KW - Parallel algorithms

KW - Performance

KW - Shared memory architectures

UR - http://www.scopus.com/inward/record.url?scp=77954896487&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=77954896487&partnerID=8YFLogxK

U2 - 10.1145/1810479.1810510

DO - 10.1145/1810479.1810510

M3 - Conference contribution

AN - SCOPUS:77954896487

SN - 9781450300797

T3 - Annual ACM Symposium on Parallelism in Algorithms and Architectures

SP - 157

EP - 165

BT - SPAA'10 - Proceedings of the 22nd Annual Symposium on Parallelism in Algorithms and Architectures

T2 - 22nd ACM Symposium on Parallelism in Algorithms and Architectures, SPAA'10

Y2 - 13 June 2010 through 15 June 2010

ER -