Using lightweight virtual machines to run high performance computing applications: The case of the weather research and forecasting model

Hector A. Duran-Limon, Luis A. Silva-Bañuelos, Victor H. Tellez-Valdez, Nikos Parlavantzas, Ming Zhao

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

5 Citations (Scopus)

Abstract

There are many scientific applications that have high performance computing demands. Such demands are traditionally supported by cluster-or Grid-based systems. Cloud computing, which has experienced a tremendous growth, emerged as an approach to provide on-demand access to computing resources. The cloud computing paradigm offers a number of advantages over other distributed platforms. For example, the access to resources is flexible and cost-effective since it is not necessary to invest a large amount of money on a computing infrastructure nor pay salaries for maintenance functions. Therefore, the possibility of using cloud computing for running high performance computing applications is attractive. However, it has been shown elsewhere that current cloud computing platforms are not suitable for running this kind of applications since the performance offered is very poor. The reason is mainly the overhead from virtualisation which is extensively used by most cloud computing platforms as a means to optimise resource usage. In this paper, we present a lightweight virtualisation approach applied to WRF, a challenging communication-intensive, high performance computing application. Our experimental results show that lightweight virtualisation imposes about 5% overhead and it substantially outperforms traditional heavy-weight virtualisation such as VMware.

Original languageEnglish (US)
Title of host publicationProceedings - 2011 4th IEEE International Conference on Utility and Cloud Computing, UCC 2011
Pages146-153
Number of pages8
DOIs
StatePublished - 2011
Externally publishedYes
Event4th IEEE/ACM International Conference on Cloud and Utility Computing, UCC 2011 - Melbourne, VIC, Australia
Duration: Dec 5 2011Dec 8 2011

Other

Other4th IEEE/ACM International Conference on Cloud and Utility Computing, UCC 2011
CountryAustralia
CityMelbourne, VIC
Period12/5/1112/8/11

Fingerprint

Cloud computing
Wages
Virtual machine
Virtualization
Communication
Costs

Keywords

  • cloud computing
  • high performance computing
  • lightweight virtual machines
  • virtualisation

ASJC Scopus subject areas

  • Software

Cite this

Duran-Limon, H. A., Silva-Bañuelos, L. A., Tellez-Valdez, V. H., Parlavantzas, N., & Zhao, M. (2011). Using lightweight virtual machines to run high performance computing applications: The case of the weather research and forecasting model. In Proceedings - 2011 4th IEEE International Conference on Utility and Cloud Computing, UCC 2011 (pp. 146-153). [6123492] https://doi.org/10.1109/UCC.2011.29

Using lightweight virtual machines to run high performance computing applications : The case of the weather research and forecasting model. / Duran-Limon, Hector A.; Silva-Bañuelos, Luis A.; Tellez-Valdez, Victor H.; Parlavantzas, Nikos; Zhao, Ming.

Proceedings - 2011 4th IEEE International Conference on Utility and Cloud Computing, UCC 2011. 2011. p. 146-153 6123492.

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

Duran-Limon, HA, Silva-Bañuelos, LA, Tellez-Valdez, VH, Parlavantzas, N & Zhao, M 2011, Using lightweight virtual machines to run high performance computing applications: The case of the weather research and forecasting model. in Proceedings - 2011 4th IEEE International Conference on Utility and Cloud Computing, UCC 2011., 6123492, pp. 146-153, 4th IEEE/ACM International Conference on Cloud and Utility Computing, UCC 2011, Melbourne, VIC, Australia, 12/5/11. https://doi.org/10.1109/UCC.2011.29
Duran-Limon HA, Silva-Bañuelos LA, Tellez-Valdez VH, Parlavantzas N, Zhao M. Using lightweight virtual machines to run high performance computing applications: The case of the weather research and forecasting model. In Proceedings - 2011 4th IEEE International Conference on Utility and Cloud Computing, UCC 2011. 2011. p. 146-153. 6123492 https://doi.org/10.1109/UCC.2011.29
Duran-Limon, Hector A. ; Silva-Bañuelos, Luis A. ; Tellez-Valdez, Victor H. ; Parlavantzas, Nikos ; Zhao, Ming. / Using lightweight virtual machines to run high performance computing applications : The case of the weather research and forecasting model. Proceedings - 2011 4th IEEE International Conference on Utility and Cloud Computing, UCC 2011. 2011. pp. 146-153
@inproceedings{89c1b8f33824477596a28287c0daa0c9,
title = "Using lightweight virtual machines to run high performance computing applications: The case of the weather research and forecasting model",
abstract = "There are many scientific applications that have high performance computing demands. Such demands are traditionally supported by cluster-or Grid-based systems. Cloud computing, which has experienced a tremendous growth, emerged as an approach to provide on-demand access to computing resources. The cloud computing paradigm offers a number of advantages over other distributed platforms. For example, the access to resources is flexible and cost-effective since it is not necessary to invest a large amount of money on a computing infrastructure nor pay salaries for maintenance functions. Therefore, the possibility of using cloud computing for running high performance computing applications is attractive. However, it has been shown elsewhere that current cloud computing platforms are not suitable for running this kind of applications since the performance offered is very poor. The reason is mainly the overhead from virtualisation which is extensively used by most cloud computing platforms as a means to optimise resource usage. In this paper, we present a lightweight virtualisation approach applied to WRF, a challenging communication-intensive, high performance computing application. Our experimental results show that lightweight virtualisation imposes about 5{\%} overhead and it substantially outperforms traditional heavy-weight virtualisation such as VMware.",
keywords = "cloud computing, high performance computing, lightweight virtual machines, virtualisation",
author = "Duran-Limon, {Hector A.} and Silva-Ba{\~n}uelos, {Luis A.} and Tellez-Valdez, {Victor H.} and Nikos Parlavantzas and Ming Zhao",
year = "2011",
doi = "10.1109/UCC.2011.29",
language = "English (US)",
isbn = "9780769545929",
pages = "146--153",
booktitle = "Proceedings - 2011 4th IEEE International Conference on Utility and Cloud Computing, UCC 2011",

}

TY - GEN

T1 - Using lightweight virtual machines to run high performance computing applications

T2 - The case of the weather research and forecasting model

AU - Duran-Limon, Hector A.

AU - Silva-Bañuelos, Luis A.

AU - Tellez-Valdez, Victor H.

AU - Parlavantzas, Nikos

AU - Zhao, Ming

PY - 2011

Y1 - 2011

N2 - There are many scientific applications that have high performance computing demands. Such demands are traditionally supported by cluster-or Grid-based systems. Cloud computing, which has experienced a tremendous growth, emerged as an approach to provide on-demand access to computing resources. The cloud computing paradigm offers a number of advantages over other distributed platforms. For example, the access to resources is flexible and cost-effective since it is not necessary to invest a large amount of money on a computing infrastructure nor pay salaries for maintenance functions. Therefore, the possibility of using cloud computing for running high performance computing applications is attractive. However, it has been shown elsewhere that current cloud computing platforms are not suitable for running this kind of applications since the performance offered is very poor. The reason is mainly the overhead from virtualisation which is extensively used by most cloud computing platforms as a means to optimise resource usage. In this paper, we present a lightweight virtualisation approach applied to WRF, a challenging communication-intensive, high performance computing application. Our experimental results show that lightweight virtualisation imposes about 5% overhead and it substantially outperforms traditional heavy-weight virtualisation such as VMware.

AB - There are many scientific applications that have high performance computing demands. Such demands are traditionally supported by cluster-or Grid-based systems. Cloud computing, which has experienced a tremendous growth, emerged as an approach to provide on-demand access to computing resources. The cloud computing paradigm offers a number of advantages over other distributed platforms. For example, the access to resources is flexible and cost-effective since it is not necessary to invest a large amount of money on a computing infrastructure nor pay salaries for maintenance functions. Therefore, the possibility of using cloud computing for running high performance computing applications is attractive. However, it has been shown elsewhere that current cloud computing platforms are not suitable for running this kind of applications since the performance offered is very poor. The reason is mainly the overhead from virtualisation which is extensively used by most cloud computing platforms as a means to optimise resource usage. In this paper, we present a lightweight virtualisation approach applied to WRF, a challenging communication-intensive, high performance computing application. Our experimental results show that lightweight virtualisation imposes about 5% overhead and it substantially outperforms traditional heavy-weight virtualisation such as VMware.

KW - cloud computing

KW - high performance computing

KW - lightweight virtual machines

KW - virtualisation

UR - http://www.scopus.com/inward/record.url?scp=84856324497&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=84856324497&partnerID=8YFLogxK

U2 - 10.1109/UCC.2011.29

DO - 10.1109/UCC.2011.29

M3 - Conference contribution

AN - SCOPUS:84856324497

SN - 9780769545929

SP - 146

EP - 153

BT - Proceedings - 2011 4th IEEE International Conference on Utility and Cloud Computing, UCC 2011

ER -