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:20230831T095745Z
LOCATION:Flüela
DTSTART;TZID=Europe/Stockholm:20230626T173000
DTEND;TZID=Europe/Stockholm:20230626T180000
UID:submissions.pasc-conference.org_PASC23_sess142_msa107@linklings.com
SUMMARY:Performance Portability Using a Tool-Chain to Exploit Separation o
 f Concerns
DESCRIPTION:Minisymposium\n\nAnshu Dubey (Argonne National Laboratory, Uni
 versity of Chicago); Jared O'Neal (Argonne National Laboratory); Johann Ru
 di (Virginia Tech, Argonne National Laboratory); Mohamed Wahib (RIKEN); To
 m Klosterman (Argonne National Laboratory); and Klaus Weide (University of
  Chicago, Argonne National Laboratory)\n\nMulti-scale, multi-physics scien
 tific and engineering simulation codes take years to develop and optimize.
  At the same time effective utilization of high performance computing (HPC
 ) resources has always been a balancing act between portability and perfor
 mance. Increasing heterogeneity and ongoing deep architecture changes in t
 he HPC platforms from one generation to the next further complicate this b
 alancing act. Abstractions within the applications codes that could utiliz
 e an ecosystem of tools such as unifiable expression of arithmetic, code t
 ransformations and runtime systems provide one possible approach that coul
 d deliver portability without significant performance loss. This presentat
 ion will outline one such approach which is applicable to a class of scien
 tific codes that solve partial differential equations with predominantly e
 xplicit methods. The tool-chain was developed for Flash-X, a composable co
 de that serves multiple domains. Fortran is the dominant language for the 
 code, which precluded the use of C++ based performance portability solutio
 ns. Our tool-chain exploits separation of concerns to keep individual tool
 s simple while being able to deliver an effective performance portability 
 solution.\n\nDomain: Computer Science, Machine Learning, and Applied Mathe
 matics 
\n\nSession Chair: Anshu Dubey (Argonne National Laboratory, Unive
 rsity of Chicago)
END:VEVENT
END:VCALENDAR