BEGIN:VCALENDAR
VERSION:2.0
PRODID:Linklings LLC
BEGIN:VTIMEZONE
TZID:America/New_York
X-LIC-LOCATION:America/New_York
BEGIN:DAYLIGHT
TZOFFSETFROM:-0500
TZOFFSETTO:-0400
TZNAME:EDT
DTSTART:19700308T020000
RRULE:FREQ=YEARLY;BYMONTH=3;BYDAY=2SU
END:DAYLIGHT
BEGIN:STANDARD
TZOFFSETFROM:-0400
TZOFFSETTO:-0500
TZNAME:EST
DTSTART:19701101T020000
RRULE:FREQ=YEARLY;BYMONTH=11;BYDAY=1SU
END:STANDARD
END:VTIMEZONE
BEGIN:VEVENT
DTSTAMP:20260422T143140Z
LOCATION:B312
DTSTART;TZID=America/New_York:20241117T140000
DTEND;TZID=America/New_York:20241117T143000
UID:submissions.supercomputing.org_SC24_sess739_ws_ss114@linklings.com
SUMMARY:Exploring the Frontiers of Energy Efficiency using Power Managemen
 t at System Scale
DESCRIPTION:Ahmad Maroof Karimi, Matthias Maiterth, Woong Shin, Naw Safrin
  Sattar, Hao Lu, and Feiyi Wang (Oak Ridge National Laboratory (ORNL))\n\n
 In the face of surging power demands for exascale HPC systems, this work t
 ackles the critical challenge of understanding the impact of software-driv
 en power management\ntechniques like Dynamic Voltage and Frequency Scaling
  (DVFS) and Power Capping. These techniques have been actively developed o
 ver the past few decades. By combining insights from\nGPU benchmarking to 
 understand application power profiles, we present a telemetry data-driven 
 approach for deriving energy savings projections. This approach has been d
 emonstrably applied\nto the Frontier supercomputer at scale. Our findings 
 based on three months of telemetry data indicate that, for certain resourc
 e constrained jobs, significant energy savings (up to 8.5%) can be\nachiev
 ed without compromising performance. This translates to a substantial cost
  reduction, equivalent to 1438 MWh of energy saved. The key contribution o
 f this work lies in the methodology for establishing an upper limit for th
 ese best-case scenarios and its successful application.\n\nTag: Energy Eff
 iciency, HPC Infrastructure, Sustainability\n\nRegistration Category: Work
 shop Reg Pass\n\nSession Chairs: Cate Berard (US Department of Energy); Ja
 mes H. Rogers (Oak Ridge National Laboratory (ORNL)); Fumiyoshi Shoji (RIK
 EN Center for Computational Science (R-CCS), Center for Computational Scie
 nce); Michèle Weiland (EPCC, The University of Edinburgh; The University o
 f Edinburgh); and Mike Woodacre (Hewlett Packard Enterprise (HPE))\n\n
END:VEVENT
END:VCALENDAR