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:20230124T171524Z
LOCATION:C155
DTSTART;TZID=America/Chicago:20221114T083100
DTEND;TZID=America/Chicago:20221114T090000
UID:submissions.supercomputing.org_SC22_sess453_ws_pmbsf119@linklings.com
SUMMARY:ML-Based Performance Portability for Time-Dependent Density Functi
 onal Theory in HPC Environments
DESCRIPTION:Workshop\n\nML-Based Performance Portability for Time-Dependen
 t Density Functional Theory in HPC Environments\n\nPerez Dieguez, Ibrahim,
  Choi, Wong, Zhu\n\nTime-Dependent Density Functional Theory (TDDFT) workl
 oads are an example of high-impact computational methods that require leve
 raging the performance of HPC architectures. However, finding the optimal 
 values of their performance-critical parameters raises performance portabi
 lity challenges that must be addressed. In this work, we propose an ML-bas
 ed tuning methodology based on Bayesian optimization and transfer learning
  to tackle the performance portability for TDDFT codes in HPC systems. Our
  results demonstrate the effectiveness of our transfer-learning proposal f
 or TDDFT workloads, which reduced the number of executed evaluations by up
  to 86%  compared to an exhaustive search for the global optimal performan
 ce parameters on the Cori and Perlmutter supercomputers. Compared to a Bay
 esian-optimization search, our proposal reduces the required evaluations b
 y up to 46.7% to find the same optimal runtime configuration. Overall, thi
 s methodology can be applied to other scientific workloads for current and
  emerging high-performance architectures.\n\nSession Format: Recorded\n\nT
 ag: Applications, Architectures, Benchmarking, Exascale Computing, Modelin
 g and Simulation, Performance, Performance Portability\n\nRegistration Cat
 egory: Workshop Reg Pass
END:VEVENT
END:VCALENDAR