BEGIN:VCALENDAR
VERSION:2.0
PRODID:Linklings LLC
BEGIN:VTIMEZONE
TZID:America/Chicago
X-LIC-LOCATION:America/Chicago
BEGIN:DAYLIGHT
TZOFFSETFROM:-0600
TZOFFSETTO:-0500
TZNAME:CDT
DTSTART:19700308T020000
RRULE:FREQ=YEARLY;BYMONTH=3;BYDAY=2SU
END:DAYLIGHT
BEGIN:STANDARD
TZOFFSETFROM:-0500
TZOFFSETTO:-0600
TZNAME:CST
DTSTART:19701101T020000
RRULE:FREQ=YEARLY;BYMONTH=11;BYDAY=1SU
END:STANDARD
END:VTIMEZONE
BEGIN:VEVENT
DTSTAMP:20181221T160729Z
LOCATION:C141/143/149
DTSTART;TZID=America/Chicago:20181113T140000
DTEND;TZID=America/Chicago:20181113T143000
UID:submissions.supercomputing.org_SC18_sess209_pap186@linklings.com
SUMMARY:Energy Efficiency Modeling of Parallel Applications
DESCRIPTION:Paper\nOpenMP, Performance, Power, Tools, Tech Program Reg Pas
 s\n\nEnergy Efficiency Modeling of Parallel Applications\n\nEndrei, Jin, D
 inh, Abramson, Poxon...\n\nEnergy efficiency has become increasingly impor
 tant in high performance computing (HPC), as power constraints and costs e
 scalate. Workload and system characteristics form a complex optimization s
 earch space in which optimal settings for energy efficiency and performanc
 e often diverge. Thus, we must identify trade-off options to find the desi
 red balance. We present an innovative statistical model that accurately pr
 edicts the Pareto optimal trade-off options using only user-controllable p
 arameters. Our approach can also tolerate both measurement and model error
 s. We study model training and validation using several HPC kernels, then 
 with more complex workloads, including AMG and LAMMPS. We can calibrate an
  accurate model from as few as 12 runs, with prediction error of less than
  10%. Our results identify trade-off options allowing up to 40% energy eff
 iciency improvement at the cost of under 20% performance loss. For AMG, we
  reduce the required sample measurement time from 13 hours to 74 minutes.
URL:https://sc18.supercomputing.org/presentation/?id=pap186&sess=sess209
END:VEVENT
END:VCALENDAR