Computational Technology Resources - CCP

Keywords: monitoring, management, HPC, energy, consumption, optimisation, adaptation, infrastructure.

Summary

Our proposed solution being framed within the EU research project GAMES [1] therefore aims to minimize the environmental burden and maximize energy efficiency without affecting the functionality and performance of the corrsponding IT service infrastructures [2]. The basis therefore has been defined with the "layered green performance indicators" [3]. The green performance indicators define a set of metrics allowing for the analysis of the energy efficiency of an IT service centre at various levels. The measurement of these corresponding metrics allows for an analysis of energy losses in applications (e.g. allowing for the determination of energy waste by deploying a service with an non-optimal configuration) and on actions that can be undertaken to save energy, such as redundancy elimination from applications, or better exploitation of middleware and processing infrastructures. However, the profitable use of green performance indicators can only be accomplished when the measurement and analysis of these metrics happens in an energy aware and efficient way. The usual measurement of IT resource usage happens by polling every single compute node cyclical for these corresponding metrics such as CPU or memory usage, so we introduced a new concept allowing for a hardware-based, fine granular monitoring of the corresponding infrastructures. This proposed, hardware-based monitoring approach allows for the monitoring of complex IT-infrastructures in a fine-granular way, whilst keeping the corresponding payload for the IT-infrastructure as small as possible. In combination with the introduced management infrastructure this approach allows for the real-time, fine-granular monitoring and adaptation of complex IT-infrastructures

References

1: GAMES project website, http://www.green-datacenters.eu/
2: M. Bullock, "Data Center Definition and Solutions", CIO, 2009.
3: "GAMES Report D2.1 - Layered Green Performance Indicators Definition", http://www.green-datacenters.eu/

purchase the full-text of this paper (price £20)

go to the previous paper
go to the next paper
return to the table of contents
return to the book description
purchase this book (price £85 +P&P)

	Computational & Technology Resources an online resource for computational, engineering & technology publications
	not logged in - login
Front Page Browse CCP CSETS CTR IJRT Other Authors Search Purchase Guide FAQ Contact us	Civil-Comp Proceedings ISSN 1759-3433 CCP: 95 PROCEEDINGS OF THE SECOND INTERNATIONAL CONFERENCE ON PARALLEL, DISTRIBUTED, GRID AND CLOUD COMPUTING FOR ENGINEERING Edited by: Paper 33 Energy Consumption Optimisation in HPC Service Centres A. Kipp¹, L. Schubert¹, J. Liu¹, T. Jiang¹, W. Christmann^2,3 and M. vor dem Berge² ¹High Performance Computing Centre Stuttgart, Germany ²Christmann Informationstechnik, Ilsede, Germany ³Medien GmbH & Co. KG, Ilsede, Germany doi:10.4203/ccp.95.33 purchase the full-text of this paper Full Bibliographic Reference for this paper A. Kipp, L. Schubert, J. Liu, T. Jiang, W. Christmann, M. vor dem Berge, "Energy Consumption Optimisation in HPC Service Centres", in , (Editors), "Proceedings of the Second International Conference on Parallel, Distributed, Grid and Cloud Computing for Engineering", Civil-Comp Press, Stirlingshire, UK, Paper 33, 2011. doi:10.4203/ccp.95.33 Keywords: monitoring, management, HPC, energy, consumption, optimisation, adaptation, infrastructure. Summary Our proposed solution being framed within the EU research project GAMES [1] therefore aims to minimize the environmental burden and maximize energy efficiency without affecting the functionality and performance of the corrsponding IT service infrastructures [2]. The basis therefore has been defined with the "layered green performance indicators" [3]. The green performance indicators define a set of metrics allowing for the analysis of the energy efficiency of an IT service centre at various levels. The measurement of these corresponding metrics allows for an analysis of energy losses in applications (e.g. allowing for the determination of energy waste by deploying a service with an non-optimal configuration) and on actions that can be undertaken to save energy, such as redundancy elimination from applications, or better exploitation of middleware and processing infrastructures. However, the profitable use of green performance indicators can only be accomplished when the measurement and analysis of these metrics happens in an energy aware and efficient way. The usual measurement of IT resource usage happens by polling every single compute node cyclical for these corresponding metrics such as CPU or memory usage, so we introduced a new concept allowing for a hardware-based, fine granular monitoring of the corresponding infrastructures. This proposed, hardware-based monitoring approach allows for the monitoring of complex IT-infrastructures in a fine-granular way, whilst keeping the corresponding payload for the IT-infrastructure as small as possible. In combination with the introduced management infrastructure this approach allows for the real-time, fine-granular monitoring and adaptation of complex IT-infrastructures References 1 GAMES project website, http://www.green-datacenters.eu/ 2 M. Bullock, "Data Center Definition and Solutions", CIO, 2009. 3 "GAMES Report D2.1 - Layered Green Performance Indicators Definition", http://www.green-datacenters.eu/ purchase the full-text of this paper (price £20) go to the previous paper go to the next paper return to the table of contents return to the book description purchase this book (price £85 +P&P)
Back to top	©Civil-Comp Limited 2023 - terms & conditions