|
|||||
|
|
Runtime power usage estimation of HPC servers for various classes of real-life applications |
|
| Affiliation: | 1. Poznan Supercomputing and Networking Center, ul. Noskowskiego 10, 61-704 Poznan, Poland;2. DEIB – Politecnico di Milano, Piazza Leonardo da Vinci 32, Milano, Italy;3. XLAB d.o.o., Pot za Brdom 100, 1000 Ljubljana, Slovenia;4. Cognitronics and Sensor Systems Group, CITEC, Bielefeld University, Germany;5. christmann informationstechnik + medien GmbH & Co. KG, Ilseder Huette 10c, 31241 Ilsede, Germany;6. CEA, LIST, PC 172, 91191 Gif-sur-Yvette CEDEX, France;7. OFFIS e.V. – Institute for Information Technology, Germany;8. ARM Ltd., CPC-1 Capital Park, Fulbourn, Cambridge CB21 5XE, UK;9. Vodafone Telematics, India;10. CEWE Stiftung & Co. KGaA, Germany |
| Abstract: | In this paper we propose a methodology underlying a development of system-wide energy consumption models for servers, which is based on the analysis of performance counters. It enables to estimate the power usage of a machine under any load at runtime. By clustering applications we extract groups of programs having similar characteristics. This allows us to create more specialized and accurate power usage models. By using decision trees it is possible to automatically select an appropriate model to current system load. Training and test sets of programs were used to test the estimates. The presented models are accurate within an error of 4% as verified on servers from different vendors, including the latest pre-production one. |
| Keywords: | Power usage estimation Application monitoring Application classification High performance computing servers |
| 本文献已被 ScienceDirect 等数据库收录！ | |