Towards an energy-aware scientific I/O interface
- 132 Downloads
Intelligently switching energy saving modes of CPUs, NICs and disks is mandatory to reduce the energy consumption.
Hardware and operating system have a limited perspective of future performance demands, thus automatic control is suboptimal. However, it is tedious for a developer to control the hardware by himself.
In this paper we propose an extension of an existing I/O interface which on the one hand is easy to use and on the other hand could steer energy saving modes more efficiently. Furthermore, the proposed modifications are beneficial for performance analysis and provide even more information to the I/O library to improve performance.
When a user annotates the program with the proposed interface, I/O, communication and computation phases are labeled by the developer. Run-time behavior is then characterized for each phase, this knowledge could be then exploited by the new library.
KeywordsScientific I/O API Energy efficiency ADIOS Performance analysis Performance optimization
Unable to display preview. Download preview PDF.
- 1.Burtscher M, Kim BD, Diamond J, McCalpin J, Koesterke L, Browne J (2010) Perfexpert: An easy-to-use performance diagnosis tool for HPC applications. In: Proceedings of the 2010 ACM/IEEE international conference for high performance computing, networking, storage and analysis, SC ’10. IEEE Computer Society, Washington, DC, pp 1–11. doi: 10.1109/SC.2010.41 CrossRefGoogle Scholar
- 2.Freeh V, Lowenthal D, Pan F, Kappiah N, Springer R, Rountree B, Femal M (2007) Analyzing the energy-time trade-off in high-performance computing applications. IEEE Trans Parallel Distrib Syst 8:1575–1590 Google Scholar
- 4.Geimer M, Wolf F, Wylie BJN, Abraham E, Becker D, Mohr B (2010) The Scalasca performance toolset architecture. Concurr Comput 22(6):277–288 Google Scholar
- 6.Hotta Y, Sato M, Kimura H, Matsuoka S, Boku T, Takahashi D (2006) Profile-based optimization of power performance by using dynamic voltage scaling on a PC cluster. In: IPDPS ’06: proceedings of the 20th international parallel and distributed processing symposium (2006). doi: 10.1109/IPDPS.2006.1639597 Google Scholar
- 9.Knüpfer A, Brunst H, Doleschal J, Jurenz M, Lieber M, Mickler H, Müller MS, Nagel WE (2008) The Vampir performance analysis tool-set. In: Tools for high performance computing, proceedings of the 2nd international workshop on parallel tools. Springer, Berlin, pp 139–155 Google Scholar
- 10.Lofstead J, Klasky SKS, Podhorszki N, Jin C (2008) Flexible IO and integration for scientific codes through the adaptable IO system (ADIOS). http://www.adiosapi.org/uploads/clade110-lofstead.pdf
- 11.Lofstead J, Zheng F, Klasky S, Schwan K (2009) Adaptable, metadata rich IO methods for portable high performance IO. In: Proceedings of IPDPS’09, May 25–29, Rome, Italy. Springer, Berlin Google Scholar
- 12.Minartz T, Knobloch M, Ludwig T, Mohr B (2011, will be published) Managing hardware power saving modes for high performance computing Google Scholar
- 14.Minartz T, Molka D, Knobloch M, Krempel S, Ludwig T, Nagel W, Mohr B, Falter H (2011, will be published) eeClust—Energy-efficient cluster computing Google Scholar
- 15.Minh TN, Wolters L (2010) Using historical data to predict application runtimes on backfilling parallel systems. In: Euromicro conference on parallel, distributed, and network-based processing, pp 246–252. http://doi.ieeecomputersociety.org/10.1109/PDP.2010.18 CrossRefGoogle Scholar
- 18.Smith W, Foster IT, Taylor VE (1998) Predicting application run times using historical information. In: Proceedings of the workshop on job scheduling strategies for parallel processing. Springer, London, pp 122–142. http://portal.acm.org/citation.cfm?id=646379.689526 CrossRefGoogle Scholar