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:20260422T000713Z
LOCATION:507
DTSTART;TZID=America/Denver:20231113T110000
DTEND;TZID=America/Denver:20231113T112000
UID:submissions.supercomputing.org_SC23_sess444_ws_waccpd109@linklings.com
SUMMARY:Performance-Portable GPU Acceleration of the EFIT Tokamak Plasma E
 quilibrium Reconstruction Code
DESCRIPTION:Oscar Antepara and Samuel Williams (Lawrence Berkeley National
  Laboratory (LBNL)); Scott Kruger (Tech-X Corporation); and Torrin Bechtel
 , Joseph McClenaghan, and Lang Lao (General Atomics)\n\nWe present the ste
 ps followed to GPU-offload parts of the core solver of EFIT-AI, an equilib
 rium reconstruction code suitable for tokamak experiments and burning plas
 mas.  For this work, we will focus on the fitting procedure that consists 
 of a Grad–Shafranov (GS) equation inverse solver that calculates equilibri
 um reconstructions on a grid. We will show profiling results of the origin
 al code(CPU-baseline), as well as the directives used to GPU-offload the m
 ost time-consuming function, initially to compare OpenACC and OpenMP on NV
 IDIA and AMD GPUs and later on to assess OpenMP performance portability on
  NVIDIA, AMD and Intel GPUs. We will make a performance comparison for dif
 ferent grid sizes and show the speedup achieved on NVIDIA A100 (Perlmutter
 -NERSC), AMD MI250X (Frontier-OLCF) and Intel PVC GPUs (Sunspot-ALCF). Fin
 ally, we will draw some conclusions and recommendations to achieve high-pe
 rformance portability for an equilibrium reconstruction code on the new HP
 C architectures.\n\nTag: Accelerators, Artificial Intelligence/Machine Lea
 rning, Algorithms, Applications, Compilers, Data Movement and Memory, Hete
 rogeneous Computing, Modeling and Simulation, Performance Optimization, Pr
 ogramming Frameworks and System Software, Runtime Systems\n\nRegistration 
 Category: Workshop Reg Pass\n\nSession Chairs: Maciej Cytowski (Pawsey Sup
 ercomputing Research Centre; Commonwealth Scientific and Industrial Resear
 ch Organisation (CSIRO), Australia); Verónica G. Melesse Vergara (Oak Ridg
 e National Laboratory (ORNL)); and Jose Manuel Monsalve Diaz (Advanced Mic
 ro Devices (AMD))\n\n
END:VEVENT
END:VCALENDAR