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:605
DTSTART;TZID=America/Denver:20231113T120300
DTEND;TZID=America/Denver:20231113T121200
UID:submissions.supercomputing.org_SC23_sess441_ws_p3hpc118@linklings.com
SUMMARY:High-Level GPU Code:  A Case Study Examining JAX and OpenMP
DESCRIPTION:Nestor Demeure, Theodore Kisner, Reijo Keskitalo, Rollin Thoma
 s, Julian Borrill, and Wahid Bhimji (Lawrence Berkeley National Laboratory
  (LBNL))\n\nIn recent years, a new scientific software design pattern has 
 emerged that pairs a Python interface with high-performance kernels in low
 er-level languages. The rise of general-purpose GPUs necessitates the rewr
 iting of many such kernels, posing challenges in GPU programming and ensur
 ing future portability and flexibility.\n\nThis paper investigates the use
  of high-level frameworks that abstract system architecture details, aimin
 g for straightforward, portable yet performant GPU code. We focus on TOAST
 , a cosmology software framework designed to take full advantage of a supe
 rcomputer, and compare using the JAX Python library with OpenMP target off
 load compiler directives as porting strategies. While JAX allows kernel co
 de to be written in pure Python, OpenMP target offload is a directive-base
 d strategy that integrates seamlessly with our existing OpenMP-accelerated
  C++ kernels.\n\nWe port a dozen kernels, analyzing development cost, perf
 ormance, and the viability of using either framework for complex numerical
  Python applications.\n\nTag: Performance Measurement, Modeling, and Tools
 , Performance Optimization\n\nRegistration Category: Workshop Reg Pass\n\n
 Session Chairs: Judith C. Hill (Lawrence Livermore National Laboratory (LL
 NL)), CJ Newburn (NVIDIA Corporation), Scott J. Parker (Argonne National L
 aboratory (ANL)), and John Pennycook (Intel Corporation)\n\n
END:VEVENT
END:VCALENDAR