Exploring hardware overprovisioning in power-constrained, high performance computing

Tapasya Patki, David K Lowenthal, Barry Rountree, Martin Schulz, Bronis R. De Supinski

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

86 Citations (Scopus)

Abstract

Most recent research in power-aware supercomputing has focused on making individual nodes more efficient and measuring the results in terms of flops per watt. While this work is vital in order to reach exascale computing at 20 megawatts, there has been a dearth of work that explores efficiency at the whole system level. Traditional approaches in supercomputer design use worst-case power provisioning: the total power allocated to the system is determined by the maximum power draw possible per node. In a world where power is plentiful and nodes are scarce, this solution is optimal. However, as power becomes the limiting factor in supercomputer design, worst-case provisioning becomes a drag on performance. In this paper we demonstrate how a policy of overprovisioning hardware with respect to power combined with intelligent, hardware-enforced power bounds consistently leads to greater performance across a range of standard benchmarks. In particular, leveraging overprovisioning requires that applications use effective configurations; the best configuration depends on application scalability and memory contention. We show that using overprovisioning leads to an average speedup of more than 50% over worst-case provisioning.

Original languageEnglish (US)
Title of host publicationProceedings of the International Conference on Supercomputing
Pages173-182
Number of pages10
DOIs
StatePublished - 2013
Event27th ACM International Conference on Supercomputing, ICS 2013 - Eugene, OR, United States
Duration: Jun 10 2013Jun 14 2013

Other

Other27th ACM International Conference on Supercomputing, ICS 2013
CountryUnited States
CityEugene, OR
Period6/10/136/14/13

Fingerprint

Supercomputers
Hardware
Drag
Scalability
Data storage equipment

Keywords

  • high-performance computing
  • overprovisioned
  • power
  • rapl

ASJC Scopus subject areas

  • Computer Science(all)

Cite this

Patki, T., Lowenthal, D. K., Rountree, B., Schulz, M., & De Supinski, B. R. (2013). Exploring hardware overprovisioning in power-constrained, high performance computing. In Proceedings of the International Conference on Supercomputing (pp. 173-182) https://doi.org/10.1145/2464996.2465009

Exploring hardware overprovisioning in power-constrained, high performance computing. / Patki, Tapasya; Lowenthal, David K; Rountree, Barry; Schulz, Martin; De Supinski, Bronis R.

Proceedings of the International Conference on Supercomputing. 2013. p. 173-182.

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

Patki, T, Lowenthal, DK, Rountree, B, Schulz, M & De Supinski, BR 2013, Exploring hardware overprovisioning in power-constrained, high performance computing. in Proceedings of the International Conference on Supercomputing. pp. 173-182, 27th ACM International Conference on Supercomputing, ICS 2013, Eugene, OR, United States, 6/10/13. https://doi.org/10.1145/2464996.2465009
Patki T, Lowenthal DK, Rountree B, Schulz M, De Supinski BR. Exploring hardware overprovisioning in power-constrained, high performance computing. In Proceedings of the International Conference on Supercomputing. 2013. p. 173-182 https://doi.org/10.1145/2464996.2465009
Patki, Tapasya ; Lowenthal, David K ; Rountree, Barry ; Schulz, Martin ; De Supinski, Bronis R. / Exploring hardware overprovisioning in power-constrained, high performance computing. Proceedings of the International Conference on Supercomputing. 2013. pp. 173-182
@inproceedings{5b74e19332b2404e8eeb3e3fc63b17ee,
title = "Exploring hardware overprovisioning in power-constrained, high performance computing",
abstract = "Most recent research in power-aware supercomputing has focused on making individual nodes more efficient and measuring the results in terms of flops per watt. While this work is vital in order to reach exascale computing at 20 megawatts, there has been a dearth of work that explores efficiency at the whole system level. Traditional approaches in supercomputer design use worst-case power provisioning: the total power allocated to the system is determined by the maximum power draw possible per node. In a world where power is plentiful and nodes are scarce, this solution is optimal. However, as power becomes the limiting factor in supercomputer design, worst-case provisioning becomes a drag on performance. In this paper we demonstrate how a policy of overprovisioning hardware with respect to power combined with intelligent, hardware-enforced power bounds consistently leads to greater performance across a range of standard benchmarks. In particular, leveraging overprovisioning requires that applications use effective configurations; the best configuration depends on application scalability and memory contention. We show that using overprovisioning leads to an average speedup of more than 50{\%} over worst-case provisioning.",
keywords = "high-performance computing, overprovisioned, power, rapl",
author = "Tapasya Patki and Lowenthal, {David K} and Barry Rountree and Martin Schulz and {De Supinski}, {Bronis R.}",
year = "2013",
doi = "10.1145/2464996.2465009",
language = "English (US)",
isbn = "9781450321303",
pages = "173--182",
booktitle = "Proceedings of the International Conference on Supercomputing",

}

TY - GEN

T1 - Exploring hardware overprovisioning in power-constrained, high performance computing

AU - Patki, Tapasya

AU - Lowenthal, David K

AU - Rountree, Barry

AU - Schulz, Martin

AU - De Supinski, Bronis R.

PY - 2013

Y1 - 2013

N2 - Most recent research in power-aware supercomputing has focused on making individual nodes more efficient and measuring the results in terms of flops per watt. While this work is vital in order to reach exascale computing at 20 megawatts, there has been a dearth of work that explores efficiency at the whole system level. Traditional approaches in supercomputer design use worst-case power provisioning: the total power allocated to the system is determined by the maximum power draw possible per node. In a world where power is plentiful and nodes are scarce, this solution is optimal. However, as power becomes the limiting factor in supercomputer design, worst-case provisioning becomes a drag on performance. In this paper we demonstrate how a policy of overprovisioning hardware with respect to power combined with intelligent, hardware-enforced power bounds consistently leads to greater performance across a range of standard benchmarks. In particular, leveraging overprovisioning requires that applications use effective configurations; the best configuration depends on application scalability and memory contention. We show that using overprovisioning leads to an average speedup of more than 50% over worst-case provisioning.

AB - Most recent research in power-aware supercomputing has focused on making individual nodes more efficient and measuring the results in terms of flops per watt. While this work is vital in order to reach exascale computing at 20 megawatts, there has been a dearth of work that explores efficiency at the whole system level. Traditional approaches in supercomputer design use worst-case power provisioning: the total power allocated to the system is determined by the maximum power draw possible per node. In a world where power is plentiful and nodes are scarce, this solution is optimal. However, as power becomes the limiting factor in supercomputer design, worst-case provisioning becomes a drag on performance. In this paper we demonstrate how a policy of overprovisioning hardware with respect to power combined with intelligent, hardware-enforced power bounds consistently leads to greater performance across a range of standard benchmarks. In particular, leveraging overprovisioning requires that applications use effective configurations; the best configuration depends on application scalability and memory contention. We show that using overprovisioning leads to an average speedup of more than 50% over worst-case provisioning.

KW - high-performance computing

KW - overprovisioned

KW - power

KW - rapl

UR - http://www.scopus.com/inward/record.url?scp=84879812114&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=84879812114&partnerID=8YFLogxK

U2 - 10.1145/2464996.2465009

DO - 10.1145/2464996.2465009

M3 - Conference contribution

AN - SCOPUS:84879812114

SN - 9781450321303

SP - 173

EP - 182

BT - Proceedings of the International Conference on Supercomputing

ER -