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:20231113T090300
DTEND;TZID=America/Denver:20231113T092200
UID:submissions.supercomputing.org_SC23_sess441_ws_p3hpc123@linklings.com
SUMMARY:A Performance-Portable SYCL Implementation of CRK-HACC for Exascal
 e
DESCRIPTION:Esteban Rangel (Argonne National Laboratory (ANL)), John Penny
 cook (Intel Corporation), Adrian Pope and Nicholas Frontiere (Argonne Nati
 onal Laboratory (ANL)), and Zhiqiang Ma and Varsha Madananth (Intel Corpor
 ation)\n\nThe first generation of exascale systems will include a variety 
 of machine architectures, featuring GPUs from multiple vendors. As a resul
 t, many developers are interested in adopting portable programming models 
 to avoid maintaining multiple versions of their code. It is necessary to d
 ocument experiences with such programming models to assist developers in u
 nderstanding the advantages and disadvantages of different approaches.\n\n
 To this end, this work evaluates the performance portability of a SYCL imp
 lementation of a large-scale cosmology application (CRK-HACC) running on G
 PUs from three different vendors: AMD, Intel, and NVIDIA. We detail the pr
 ocess of migrating the original code from CUDA to SYCL and show that speci
 alizing kernels for specific targets can greatly improve performance porta
 bility without significantly impacting programmer productivity. The SYCL v
 ersion of CRK-HACC achieves a performance portability of 0.96 with a code 
 divergence of almost 0, demonstrating that SYCL is a viable programming mo
 del for performance-portable applications.\n\nTag: Performance Measurement
 , Modeling, and Tools, Performance Optimization\n\nRegistration Category: 
 Workshop Reg Pass\n\nSession Chairs: Judith C. Hill (Lawrence Livermore Na
 tional Laboratory (LLNL)), CJ Newburn (NVIDIA Corporation), Scott J. Parke
 r (Argonne National Laboratory (ANL)), and John Pennycook (Intel Corporati
 on)\n\n
END:VEVENT
END:VCALENDAR