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:20260624T171342Z
LOCATION:Bldg. 8 - Entrance Hall
DTSTART;TZID=Europe/Stockholm:20260630T173000
DTEND;TZID=Europe/Stockholm:20260630T194500
UID:submissions.pasc-conference.org_PASC26_sess135_pos126@linklings.com
SUMMARY:P40 - Uncertainty Quantification for Energy Efficiency Analysis of
  Scientific Applications at Exascale
DESCRIPTION:Matheus Machado, Mariana Costa, Matheus Costa, Philippe Navaux
 , and Arthur Lorenzon (UFRGS) and Antigoni Georgiadou and Bronson Messer (
 Oak Ridge National Laboratory)\n\nEnergy efficiency has become a critical 
 constraint in high-performance computing (HPC) as systems scale toward lar
 ger node counts. In modern HPC platforms, energy consumption is influenced
  by complex interactions among hardware and software parameters, including
  operating frequencies, concurrency levels, memory behavior, and runtime p
 olicies. These interactions are further affected by execution-time variabi
 lity arising from hardware heterogeneity, resource contention, and system 
 noise. As a result, energy measurements often exhibit significant fluctuat
 ions that are not captured by deterministic models or single-run experimen
 ts, limiting the reliability of traditional energy-aware optimization appr
 oaches. This work proposes using uncertainty quantification (UQ) as a syst
 ematic method for analyzing energy efficiency in scientific HPC applicatio
 ns. Instead of relying on average-case behavior, we explicitly model the v
 ariability in execution time and energy consumption as functions of system
  configuration parameters. We present a tool-based methodology that combin
 es controlled experimental measurements with variance-based sensitivity an
 alysis, surrogate modeling, and uncertainty propagation. This approach ena
 bles the identification of configuration parameters that most strongly aff
 ect energy consumption and performance variability, as well as the assessm
 ent of energy-performance trade-offs under uncertainty. Moreover, the prop
 osed tool is application-agnostic and applicable across a wide range of sc
 ientific HPC workloads and architectures.\n\n
END:VEVENT
END:VCALENDAR