BEGIN:VCALENDAR
VERSION:2.0
PRODID:Linklings LLC
BEGIN:VTIMEZONE
TZID:Europe/Stockholm
X-LIC-LOCATION:Europe/Stockholm
BEGIN:DAYLIGHT
TZOFFSETFROM:+0100
TZOFFSETTO:+0200
TZNAME:CEST
DTSTART:19700308T020000
RRULE:FREQ=YEARLY;BYMONTH=3;BYDAY=-1SU
END:DAYLIGHT
BEGIN:STANDARD
TZOFFSETFROM:+0200
TZOFFSETTO:+0100
TZNAME:CET
DTSTART:19701101T020000
RRULE:FREQ=YEARLY;BYMONTH=10;BYDAY=-1SU
END:STANDARD
END:VTIMEZONE
BEGIN:VEVENT
DTSTAMP:20230831T095747Z
LOCATION:Sanada I
DTSTART;TZID=Europe/Stockholm:20230628T143000
DTEND;TZID=Europe/Stockholm:20230628T150000
UID:submissions.pasc-conference.org_PASC23_sess138_msa163@linklings.com
SUMMARY:Going Green: Optimizing GPUs for Energy Efficiency through Model-S
 teered Auto-Tuning
DESCRIPTION:Minisymposium\n\nBram Veenboer (ASTRON) and Stefano Corda (EPF
 L)\n\nGraphics Processing Units (GPUs) have revolutionized the computing l
 andscape over the past decade. However, the growing energy demands of data
  centres and computing facilities equipped with GPUs come with significant
  capital and environmental costs. The energy consumption of GPU applicatio
 ns greatly depend on how well they are optimized. Auto-tuning is an effect
 ive and commonly applied technique of finding the optimal combination of a
 lgorithm, application, and hardware parameters to optimize performance of 
 a GPU application. In this paper, we introduce new energy monitoring and o
 ptimization capabilities in Kernel Tuner, a generic auto-tuning tool for G
 PU applications. These capabilities enable us to investigate the differenc
 e between tuning for execution time and various approaches to improve ener
 gy efficiency, and investigate the differences in tuning difficulty. Addit
 ionally, our model for GPU power consumption greatly reduces the large tun
 ing search space by providing clock frequencies for which a GPU is likely 
 most energy efficient.\n\nDomain: Computer Science, Machine Learning, and 
 Applied Mathematics 
\n\nSession Chair: Emma Tolley (EPFL)
END:VEVENT
END:VCALENDAR