A'Cool'load balancer for parallel applications

Osman Sarood, Laxmikant V Kale

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

Abstract

Meeting power requirements of huge exascale machines of the future will be a major challenge. Our focus in this paper is to minimize cooling power and we propose a technique that uses a combination of DVFS and temperature aware load balancing to constrain core temperatures as well as save cooling energy. Our scheme is specifically designed to suit parallel applications which are typically tightly coupled. The temperature control, comes at the cost of execution time and we try to minimize the timing penalty. We experiment with three applications (with different power utilization profiles), run on a 128-core (32-node) cluster with a dedicated air conditioning unit. We calibrate the efficacy of our scheme based on three metrics: ability to control average core temperatures thereby avoiding hot spot occurence, timing penalty minimization, and cooling energy savings. Our results show cooling energy savings of up to 57% with a timing penalty of 19%.

Original languageEnglish (US)
Title of host publicationProceedings of 2011 SC - International Conference for High Performance Computing, Networking, Storage and Analysis
DOIs
StatePublished - Dec 14 2011
Event2011 International Conference for High Performance Computing, Networking, Storage and Analysis, SC11 - Seattle, WA, United States
Duration: Nov 12 2011Nov 18 2011

Publication series

NameProceedings of 2011 SC - International Conference for High Performance Computing, Networking, Storage and Analysis

Other

Other2011 International Conference for High Performance Computing, Networking, Storage and Analysis, SC11
CountryUnited States
CitySeattle, WA
Period11/12/1111/18/11

Fingerprint

Cooling
Energy conservation
Temperature control
Air conditioning
Temperature
Resource allocation
Electric power utilization
Experiments

ASJC Scopus subject areas

  • Computer Networks and Communications
  • Computer Science Applications

Cite this

Sarood, O., & Kale, L. V. (2011). A'Cool'load balancer for parallel applications. In Proceedings of 2011 SC - International Conference for High Performance Computing, Networking, Storage and Analysis [21] (Proceedings of 2011 SC - International Conference for High Performance Computing, Networking, Storage and Analysis). https://doi.org/10.1145/2063384.2063412

A'Cool'load balancer for parallel applications. / Sarood, Osman; Kale, Laxmikant V.

Proceedings of 2011 SC - International Conference for High Performance Computing, Networking, Storage and Analysis. 2011. 21 (Proceedings of 2011 SC - International Conference for High Performance Computing, Networking, Storage and Analysis).

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

Sarood, O & Kale, LV 2011, A'Cool'load balancer for parallel applications. in Proceedings of 2011 SC - International Conference for High Performance Computing, Networking, Storage and Analysis., 21, Proceedings of 2011 SC - International Conference for High Performance Computing, Networking, Storage and Analysis, 2011 International Conference for High Performance Computing, Networking, Storage and Analysis, SC11, Seattle, WA, United States, 11/12/11. https://doi.org/10.1145/2063384.2063412
Sarood O, Kale LV. A'Cool'load balancer for parallel applications. In Proceedings of 2011 SC - International Conference for High Performance Computing, Networking, Storage and Analysis. 2011. 21. (Proceedings of 2011 SC - International Conference for High Performance Computing, Networking, Storage and Analysis). https://doi.org/10.1145/2063384.2063412
Sarood, Osman ; Kale, Laxmikant V. / A'Cool'load balancer for parallel applications. Proceedings of 2011 SC - International Conference for High Performance Computing, Networking, Storage and Analysis. 2011. (Proceedings of 2011 SC - International Conference for High Performance Computing, Networking, Storage and Analysis).
@inproceedings{14e2014ca8f44e99afa5259e8418c879,
title = "A'Cool'load balancer for parallel applications",
abstract = "Meeting power requirements of huge exascale machines of the future will be a major challenge. Our focus in this paper is to minimize cooling power and we propose a technique that uses a combination of DVFS and temperature aware load balancing to constrain core temperatures as well as save cooling energy. Our scheme is specifically designed to suit parallel applications which are typically tightly coupled. The temperature control, comes at the cost of execution time and we try to minimize the timing penalty. We experiment with three applications (with different power utilization profiles), run on a 128-core (32-node) cluster with a dedicated air conditioning unit. We calibrate the efficacy of our scheme based on three metrics: ability to control average core temperatures thereby avoiding hot spot occurence, timing penalty minimization, and cooling energy savings. Our results show cooling energy savings of up to 57{\%} with a timing penalty of 19{\%}.",
author = "Osman Sarood and Kale, {Laxmikant V}",
year = "2011",
month = "12",
day = "14",
doi = "10.1145/2063384.2063412",
language = "English (US)",
isbn = "9781450307710",
series = "Proceedings of 2011 SC - International Conference for High Performance Computing, Networking, Storage and Analysis",
booktitle = "Proceedings of 2011 SC - International Conference for High Performance Computing, Networking, Storage and Analysis",

}

TY - GEN

T1 - A'Cool'load balancer for parallel applications

AU - Sarood, Osman

AU - Kale, Laxmikant V

PY - 2011/12/14

Y1 - 2011/12/14

N2 - Meeting power requirements of huge exascale machines of the future will be a major challenge. Our focus in this paper is to minimize cooling power and we propose a technique that uses a combination of DVFS and temperature aware load balancing to constrain core temperatures as well as save cooling energy. Our scheme is specifically designed to suit parallel applications which are typically tightly coupled. The temperature control, comes at the cost of execution time and we try to minimize the timing penalty. We experiment with three applications (with different power utilization profiles), run on a 128-core (32-node) cluster with a dedicated air conditioning unit. We calibrate the efficacy of our scheme based on three metrics: ability to control average core temperatures thereby avoiding hot spot occurence, timing penalty minimization, and cooling energy savings. Our results show cooling energy savings of up to 57% with a timing penalty of 19%.

AB - Meeting power requirements of huge exascale machines of the future will be a major challenge. Our focus in this paper is to minimize cooling power and we propose a technique that uses a combination of DVFS and temperature aware load balancing to constrain core temperatures as well as save cooling energy. Our scheme is specifically designed to suit parallel applications which are typically tightly coupled. The temperature control, comes at the cost of execution time and we try to minimize the timing penalty. We experiment with three applications (with different power utilization profiles), run on a 128-core (32-node) cluster with a dedicated air conditioning unit. We calibrate the efficacy of our scheme based on three metrics: ability to control average core temperatures thereby avoiding hot spot occurence, timing penalty minimization, and cooling energy savings. Our results show cooling energy savings of up to 57% with a timing penalty of 19%.

UR - http://www.scopus.com/inward/record.url?scp=83155184565&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=83155184565&partnerID=8YFLogxK

U2 - 10.1145/2063384.2063412

DO - 10.1145/2063384.2063412

M3 - Conference contribution

AN - SCOPUS:83155184565

SN - 9781450307710

T3 - Proceedings of 2011 SC - International Conference for High Performance Computing, Networking, Storage and Analysis

BT - Proceedings of 2011 SC - International Conference for High Performance Computing, Networking, Storage and Analysis

ER -