EFFICIENT POWER MANAGEMENT IN HIGH PERFORMANCE COMPUTER CLUSTERS

Carlos de Alfonso Laguna, Miguel Caballer, Vicente Hernández

Abstract

Nowadays the computer cluster infrastructures are very common not only in research centers, but also in the business environment. These computer infrastructures were traditionally used to perform complex mathematic calculations using parallel programs, but currently it is extended the use of them as virtualization platforms and shared by means of grid middlewares. Anyway, this kind of infrastructures usually has many idle periods due to different issues such as the lack of intensive calculations, holydays, weekends, etc. Even having reduced the power consumption of different components, and accomplishing energy saving initiatives, it would be better to not to waste energy when it is not needed. This paper shows that green computing techniques can be applied in these infrastructures, powering-off the nodes which are not being used and switching them on when needed, to enable reducing power consumption considerably. These techniques must be applied trying to cause a minimum impact to the user who is using the infrastructure, and without interference in running applications.

References

  1. Murugesan S., 2008. Harnessing Green IT: Principles and Practices. In IEEE IT Professional, pp 24-33.
  2. Fontan, J. et al, 2008. OpenNebula: The Open Source Virtual Machine Manager for Cluster Computing. In Open Source Grid and Cluster Software Conference. San Francisco, CA, USA.
  3. Nurmi, D. et al., 2008. The Eucalyptus Open-source Cloud-computing System. In Proceedings of Cloud Computing and Its Applications.
  4. Liang, S. A., 2008. Low cost and high efficiency PC power supply design to meet 80 plus requirements. In Industrial Technology IEEE International Conference, pp 1-6.
  5. The 80 PLUS Program, 2009. viewed 1 December 2009. <http://www.80plus.org/>.
  6. Hoelzle, U., Weihl, B., 2010. High-efficiency power supplies for home computers and servers, Google Inc. viewed 22 February, 2010) <http://static. googleusercontent.com/external_content/untrusted/ser vices.google.com/blog_resources/PSU_white_paper. pdf>.
  7. von Laszewski, G. et al, 2009. Power-Aware Scheduling of Virtual Machines in DVFSenabled Clusters. In IEEE Cluster 2009. New Orleans.
  8. Feng, W. and Feng, X., 2008. Green supercomputing comes of age. IT Professional, vol. 10, no. 1, pp. 17- 23.
  9. Feng, W. and Cameron K., 2007. The Green500 List: Encouraging Sustainable Supercomputing. In IEEE Computer, pp. 50-55.
  10. Da-Costa, G. et al, 2009. The green-net framework: Energy efficiency in large scale distributed systems. In High Performance Power Aware Computing Workshop.
  11. Globus Alliance, 2009. Globus Toolkit, viewed 1 December 2009, <http://www.globus.org/toolkit>.
  12. EGEE, 2009, gLite Lightweight Middleware for Grid Computing. viewed 1 December 2009, <http://www.glite.org>
  13. Lucke, R. W., 2004. Building Clustered Linux Systems, Prentice Hall. ISBN - 978-0-13-144853-7.
Download


Paper Citation


in Harvard Style

de Alfonso Laguna C., Caballer M. and Hernández V. (2010). EFFICIENT POWER MANAGEMENT IN HIGH PERFORMANCE COMPUTER CLUSTERS . In Proceedings of the Multi-Conference on Innovative Developments in ICT - Volume 1: ICGREEN, (INNOV 2010) ISBN 978-989-8425-15-7, pages 39-44. DOI: 10.5220/0003036500390044


in Bibtex Style

@conference{icgreen10,
author={Carlos de Alfonso Laguna and Miguel Caballer and Vicente Hernández},
title={EFFICIENT POWER MANAGEMENT IN HIGH PERFORMANCE COMPUTER CLUSTERS},
booktitle={Proceedings of the Multi-Conference on Innovative Developments in ICT - Volume 1: ICGREEN, (INNOV 2010)},
year={2010},
pages={39-44},
publisher={SciTePress},
organization={INSTICC},
doi={10.5220/0003036500390044},
isbn={978-989-8425-15-7},
}


in EndNote Style

TY - CONF
JO - Proceedings of the Multi-Conference on Innovative Developments in ICT - Volume 1: ICGREEN, (INNOV 2010)
TI - EFFICIENT POWER MANAGEMENT IN HIGH PERFORMANCE COMPUTER CLUSTERS
SN - 978-989-8425-15-7
AU - de Alfonso Laguna C.
AU - Caballer M.
AU - Hernández V.
PY - 2010
SP - 39
EP - 44
DO - 10.5220/0003036500390044