- Pressemitteilung BoxID 422884
SuperMUC: Leading new Energy Efficiency Measurements
13th edition of Green500 list introduces new measurement quality levels
Based on selected quality aspects of the measurement, such as the fraction of the machine that was instrumented, granularity, timespan, and the number of included machine subsystems, a submission was rated conforming to one of three increasingly rigorous measurement quality levels. Level 1 is similar to the Green500 run rules today. Level 2 is more comprehensive and still widely achievable. Level 3 represents the "current best" in supercomputing power measurement methodology.
"Power consumption of HPC systems and subsequently facilities cost are increasing", says Natalie Bates, head of the Energy Efficient HPC Working Group that was responsible for designing the new measurement guidelines and the definition of the requirements for levels 1-3. "You can only improve what you can measure", Bates continues, "and the ability to measure power and energy will accelerate improvements in energy efficiency in supercomputing. The measurement methodology was developed as a collaborative effort between different research and industry groups." The new methodology was tested by a small group of early adopters for the June and November 2012 Green500 list releases and subsequently refined given the feedback from the early adopters.
Arndt Bode concludes: "In order to maximize the energy efficiency of the Leibniz Supercomputing Centre, a detailed understanding of the power consumption of all technical equipment needed to operate our supercomputer is required. Hence, measuring the system power consumption using high quality power measurements is only the first step of LRZ's effort in tuning our HPC infrastructure for maximum energy efficiency."
Green500 List: http://www.green500.org/
Energy Efficient HPC Working Group: http://eehpcwg.lbl.gov/
Leibniz Supercomputing Centre: http://www.lrz.de/
The Leibniz Supercomputing Centre (LRZ) of the Bavarian Academy of Sciences is the general IT service provider to the scientific and academic communities in the Munich area. LRZ operates the Munich Scientific Network (MWN), offers archiving and backup of large amounts of data and delivers high performance and supercomputing facilities for all German universities and PRACE partners via the Gauss Centre for Supercomputing (GCS).
Diese Pressemitteilungen könnten Sie auch interessieren
Die Niedersächsischen Landesforsten sind ein Hauptlieferant für übergroße Weihnachtsbäume in Deutschland. Die Ernte von insgesamt vier XXL-Fichten...
Die Naturschutzgebiete "Ziegelwerder" sowie "Kaninchenwerder und Großer Stein im Schweriner See" sind wieder offen für Wassersportler. Das Bundesministerium...