Potentials and Limitations for Energy Efficiency Auto-Tuning
Research output: Contribution to book/Conference proceedings/Anthology/Report › Chapter in book/Anthology/Report › Contributed › peer-review
Contributors
Abstract
With an objective target of 20 MW for an Exascale system, the power budget for future super computers has a severe restriction. Fortunately, current hardware and runtime environments feature various adjustable parameters that can be changed at runtime in order to confine the power usage. We present an overview of the potentials and limitations of several power saving mechanisms and describe the integration of energy efficiency optimizations in existing performance analysis tools. We also introduce an energy efficiency tuning cycle and exemplify its usage.
Details
Original language | English |
---|---|
Title of host publication | Parallel Computing: Accelerating Computational Sciene and Engineering (CSE) |
Editors | Michael Bader, Arndt Bode, Hans-Joachim Bungartz, Michael Gerndt, Gerhard R. Joubert, Frans Peters |
Publisher | IOS Press, Amsterdam [u. a.] |
Pages | 678-687 |
Number of pages | 10 |
Volume | 25 |
ISBN (print) | 978-1-61499-381-3 |
Publication status | Published - 1 Dec 2013 |
Peer-reviewed | Yes |
Conference
Title | International Conference on Parallel Computing 2013 |
---|---|
Abbreviated title | ParCo 2013 |
Duration | 10 - 13 September 2013 |
Website | |
Degree of recognition | International event |
Location | Technische Universität München |
City | München |
Country | Germany |
External IDs
Scopus | 84902284275 |
---|---|
ORCID | /0009-0003-0666-4166/work/151475604 |