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:20260202T201806Z
LOCATION:231
DTSTART;TZID=America/Chicago:20251117T160000
DTEND;TZID=America/Chicago:20251117T163000
UID:submissions.supercomputing.org_SC25_sess207_ws_p3hpc105@linklings.com
SUMMARY:Extending RAJA Parallel Programming Abstractions with Just-In-Time
  Optimization
DESCRIPTION:John Bowen, Konstantinos Parasyris, David Beckingsale, Tal Ben
 -Nun, Thomas Stitt, and Giorgis Georgakoudis (Lawrence Livermore National 
 Laboratory (LLNL))\n\nThe prevalence of heterogeneous computing systems --
  comprising both CPUs and GPUs -- has led to the adoption of performance p
 ortability programming models, such as RAJA. These models allow developers
  to write portable code that compiles ahead-of-time (AOT), unmodified for 
 different backends, thus improving productivity and maintainability. \n\nI
 n this work, we explore the integration of just-in-time (JIT) optimization
  into portable programming models. Our work aims to improve performance wi
 th JIT optimization, without sacrificing portability or developer producti
 vity. \n\nWe extend Proteus to support indirect kernel launching through R
 AJA's abstractions.  Our evaluation with the RAJAPerf benchmark suite demo
 nstrates promising speedups for both AMD and NVIDIA GPUs, with no slowdown
 s recorded for either backend.  Specifically, we record speedups from $1.2
 \times$ up to $23\times$ on AMD MI250X and speedups from $1.1\times$ up to
  $15\times$ on NVIDIA V100, while preserving the performance portability a
 nd ease-of-use benefits of RAJA.\n\nTag: Performance Evaluation, Scalabili
 ty, & Portability\n\nRecording: Livestreamed, Recorded\n\nRegistration Cat
 egory: Technical Program Reg Pass, Workshop Reg Pass\n\nSession Chairs: CJ
  Newburn (NVIDIA Corporation), Scott J. Parker (Argonne National Laborator
 y (ANL)), John Pennycook (Intel Corporation), and Kenneth Weiss (Lawrence 
 Livermore National Laboratory (LLNL))\n\n
END:VEVENT
END:VCALENDAR
