BEGIN:VCALENDAR
VERSION:2.0
PRODID:Linklings LLC
BEGIN:VTIMEZONE
TZID:America/Denver
X-LIC-LOCATION:America/Denver
BEGIN:DAYLIGHT
TZOFFSETFROM:-0700
TZOFFSETTO:-0600
TZNAME:MDT
DTSTART:19700308T020000
RRULE:FREQ=YEARLY;BYMONTH=3;BYDAY=2SU
END:DAYLIGHT
BEGIN:STANDARD
TZOFFSETFROM:-0600
TZOFFSETTO:-0700
TZNAME:MST
DTSTART:19701101T020000
RRULE:FREQ=YEARLY;BYMONTH=11;BYDAY=1SU
END:STANDARD
END:VTIMEZONE
BEGIN:VEVENT
DTSTAMP:20260422T000711Z
LOCATION:603
DTSTART;TZID=America/Denver:20231112T115500
DTEND;TZID=America/Denver:20231112T121000
UID:submissions.supercomputing.org_SC23_sess420_ws_ss112@linklings.com
SUMMARY:ReAPER: Region Aware Power and Energy Regulator
DESCRIPTION:Anna Yue, Torsten Wilde, and Steven Martin (Hewlett Packard En
 terprise (HPE)); Pen-Chung Yew (University of Minnesota, Twin Cities); and
  Sanyam Mehta (Hewlett Packard Enterprise (HPE))\n\nImproving CPU power/en
 ergy efficiency without degrading performance requires an accurate applica
 tion characterization. Rather than characterizing the application as a who
 le, we find that dividing the application into individual regions is much 
 more effective. This fine-grained approach gives us the opportunity to sav
 e power/energy during memory-bound regions and MPI slack regions (time spe
 nt waiting on other processes) by lowering core frequency and during compu
 te-bound regions by lowering uncore frequency. We propose an intuitive, li
 ghtweight, and portable algorithm for identifying these regions at runtime
  which relies only on the IPS (instructions per second) metric, rather tha
 n on performance counters that can differ across platforms. At the same ti
 me, we meet a user-specified level of acceptable performance degradation b
 y adapting core and uncore frequencies to the application, achieving addit
 ional CPU power/energy savings. We evaluate our approach on the SPEC 2017 
 benchmarks and various MPI applications.\n\nTag: Energy Efficiency, Green 
 Computing, Sustainability\n\nRegistration Category: Workshop Reg Pass\n\nS
 ession Chairs: Kimmo Koski (CSC – IT Center for Science Ltd, Finland); Jam
 es H. Rogers (Oak Ridge National Laboratory (ORNL)); Fumiyoshi Shoji (RIKE
 N Center for Computational Science (R-CCS), Center for Computational Scien
 ce); William W. Thigpen (NASA); Michèle Weiland (EPCC, The University of E
 dinburgh; The University of Edinburgh); and Mike Woodacre (Hewlett Packard
  Enterprise (HPE))\n\n
END:VEVENT
END:VCALENDAR