Communication based proactive link power management

Sai Prashanth Muralidhara, Mahmut Kandemir

Research output: Chapter in Book/Report/Conference proceedingConference contribution

Abstract

As the number of cores in CMPs increases, NoC is projected to be the dominant communication fabric. Increase in the number of cores brings an important issue to the forefront, the issue of chip power consumption, which is projected to increase rapidly with the increase in number of cores. Since NoC infrastructure contributes significantly to the total chip power consumption, reducing NoC power is crucial. While circuit level techniques are important in reducing NoC power, architectural and software level approaches can be very effective in optimizing power consumption. Any such technique power saving technique should be scalable and have minimal adverse impact on performance. We propose a dynamic, communication link usage based, proactive link power management scheme. This scheme,using a Markov model, proactively manages communication link turn-ons and turn-offs, which results in negligible performance degradation and significant power savings. We show that our prediction scheme is about 98% accurate for the SPEC OMP benchmarks and about 93% over all applications experimented. This accuracy helps us achieve link power savings of up to 44% and an average link power savings of 23.5%. More importantly, it incurs performance penalties as low as 0.3% on average.

Original languageEnglish (US)
Title of host publicationHigh Performance Embedded Architectures and Compilers - Fourth International Conference, HiPEAC 2009, Proceedings
Pages198-215
Number of pages18
DOIs
StatePublished - Feb 4 2009
Event4th International Conference on High Performance Embedded Architectures and Compilers, HiPEAC 2009 - Paphos, Cyprus
Duration: Jan 25 2009Jan 28 2009

Publication series

NameLecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)
Volume5409 LNCS
ISSN (Print)0302-9743
ISSN (Electronic)1611-3349

Other

Other4th International Conference on High Performance Embedded Architectures and Compilers, HiPEAC 2009
CountryCyprus
CityPaphos
Period1/25/091/28/09

Fingerprint

Power Saving
Power Management
Power Consumption
Telecommunication links
Electric power utilization
Communication
Chip
Markov Model
Penalty
Microprocessor chips
Degradation
Infrastructure
Benchmark
Software
Network on chip
Network-on-chip
Power management
Networks (circuits)
Prediction

All Science Journal Classification (ASJC) codes

  • Theoretical Computer Science
  • Computer Science(all)

Cite this

Muralidhara, S. P., & Kandemir, M. (2009). Communication based proactive link power management. In High Performance Embedded Architectures and Compilers - Fourth International Conference, HiPEAC 2009, Proceedings (pp. 198-215). (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics); Vol. 5409 LNCS). https://doi.org/10.1007/978-3-540-92990-1_16
Muralidhara, Sai Prashanth ; Kandemir, Mahmut. / Communication based proactive link power management. High Performance Embedded Architectures and Compilers - Fourth International Conference, HiPEAC 2009, Proceedings. 2009. pp. 198-215 (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)).
@inproceedings{559c94ee8aff41988e7bd3ec95cae1ee,
title = "Communication based proactive link power management",
abstract = "As the number of cores in CMPs increases, NoC is projected to be the dominant communication fabric. Increase in the number of cores brings an important issue to the forefront, the issue of chip power consumption, which is projected to increase rapidly with the increase in number of cores. Since NoC infrastructure contributes significantly to the total chip power consumption, reducing NoC power is crucial. While circuit level techniques are important in reducing NoC power, architectural and software level approaches can be very effective in optimizing power consumption. Any such technique power saving technique should be scalable and have minimal adverse impact on performance. We propose a dynamic, communication link usage based, proactive link power management scheme. This scheme,using a Markov model, proactively manages communication link turn-ons and turn-offs, which results in negligible performance degradation and significant power savings. We show that our prediction scheme is about 98{\%} accurate for the SPEC OMP benchmarks and about 93{\%} over all applications experimented. This accuracy helps us achieve link power savings of up to 44{\%} and an average link power savings of 23.5{\%}. More importantly, it incurs performance penalties as low as 0.3{\%} on average.",
author = "Muralidhara, {Sai Prashanth} and Mahmut Kandemir",
year = "2009",
month = "2",
day = "4",
doi = "10.1007/978-3-540-92990-1_16",
language = "English (US)",
isbn = "3540929894",
series = "Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)",
pages = "198--215",
booktitle = "High Performance Embedded Architectures and Compilers - Fourth International Conference, HiPEAC 2009, Proceedings",

}

Muralidhara, SP & Kandemir, M 2009, Communication based proactive link power management. in High Performance Embedded Architectures and Compilers - Fourth International Conference, HiPEAC 2009, Proceedings. Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics), vol. 5409 LNCS, pp. 198-215, 4th International Conference on High Performance Embedded Architectures and Compilers, HiPEAC 2009, Paphos, Cyprus, 1/25/09. https://doi.org/10.1007/978-3-540-92990-1_16

Communication based proactive link power management. / Muralidhara, Sai Prashanth; Kandemir, Mahmut.

High Performance Embedded Architectures and Compilers - Fourth International Conference, HiPEAC 2009, Proceedings. 2009. p. 198-215 (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics); Vol. 5409 LNCS).

Research output: Chapter in Book/Report/Conference proceedingConference contribution

TY - GEN

T1 - Communication based proactive link power management

AU - Muralidhara, Sai Prashanth

AU - Kandemir, Mahmut

PY - 2009/2/4

Y1 - 2009/2/4

N2 - As the number of cores in CMPs increases, NoC is projected to be the dominant communication fabric. Increase in the number of cores brings an important issue to the forefront, the issue of chip power consumption, which is projected to increase rapidly with the increase in number of cores. Since NoC infrastructure contributes significantly to the total chip power consumption, reducing NoC power is crucial. While circuit level techniques are important in reducing NoC power, architectural and software level approaches can be very effective in optimizing power consumption. Any such technique power saving technique should be scalable and have minimal adverse impact on performance. We propose a dynamic, communication link usage based, proactive link power management scheme. This scheme,using a Markov model, proactively manages communication link turn-ons and turn-offs, which results in negligible performance degradation and significant power savings. We show that our prediction scheme is about 98% accurate for the SPEC OMP benchmarks and about 93% over all applications experimented. This accuracy helps us achieve link power savings of up to 44% and an average link power savings of 23.5%. More importantly, it incurs performance penalties as low as 0.3% on average.

AB - As the number of cores in CMPs increases, NoC is projected to be the dominant communication fabric. Increase in the number of cores brings an important issue to the forefront, the issue of chip power consumption, which is projected to increase rapidly with the increase in number of cores. Since NoC infrastructure contributes significantly to the total chip power consumption, reducing NoC power is crucial. While circuit level techniques are important in reducing NoC power, architectural and software level approaches can be very effective in optimizing power consumption. Any such technique power saving technique should be scalable and have minimal adverse impact on performance. We propose a dynamic, communication link usage based, proactive link power management scheme. This scheme,using a Markov model, proactively manages communication link turn-ons and turn-offs, which results in negligible performance degradation and significant power savings. We show that our prediction scheme is about 98% accurate for the SPEC OMP benchmarks and about 93% over all applications experimented. This accuracy helps us achieve link power savings of up to 44% and an average link power savings of 23.5%. More importantly, it incurs performance penalties as low as 0.3% on average.

UR - http://www.scopus.com/inward/record.url?scp=59049088514&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=59049088514&partnerID=8YFLogxK

U2 - 10.1007/978-3-540-92990-1_16

DO - 10.1007/978-3-540-92990-1_16

M3 - Conference contribution

AN - SCOPUS:59049088514

SN - 3540929894

SN - 9783540929895

T3 - Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)

SP - 198

EP - 215

BT - High Performance Embedded Architectures and Compilers - Fourth International Conference, HiPEAC 2009, Proceedings

ER -

Muralidhara SP, Kandemir M. Communication based proactive link power management. In High Performance Embedded Architectures and Compilers - Fourth International Conference, HiPEAC 2009, Proceedings. 2009. p. 198-215. (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)). https://doi.org/10.1007/978-3-540-92990-1_16