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:20241120T082409Z
LOCATION:HG E 1.2
DTSTART;TZID=Europe/Stockholm:20240604T163000
DTEND;TZID=Europe/Stockholm:20240604T170000
UID:submissions.pasc-conference.org_PASC24_sess114_msa193@linklings.com
SUMMARY:GROMACS: Using SYCL for Performance and Portability at Scale
DESCRIPTION:Minisymposium\n\nAndrey Alekseenko (KTH Royal Institute of Tec
 hnology, SciLifeLab)\n\nMolecular dynamics simulations have quickly embrac
 ed General-Purpose computing on Graphics Processing Units. As GPU performa
 nce grew, the strong scaling problem became apparent due to the fixed size
  of typical biophysical systems.\n\nGROMACS is a widely used molecular dyn
 amics engine, designed for both high performance and portability across ha
 rdware and software. Originally, when GROMACS 2021 introduced support for 
 SYCL—an open standard for parallel computing—it was only compatible with I
 ntel GPUs. Since then, its compatibility has been significantly broadened 
 to include GPUs from all three major vendors. Although GROMACS still recom
 mends the CUDA backend for NVIDIA devices for optimal performance, SYCL is
  the recommended option for Intel and AMD GPUs.\n\nThis presentation will 
 share insights from deploying GROMACS with SYCL across both multi-GPU and 
 multi-node setups, using Intel oneAPI DPC++ and AdaptiveCpp runtimes. We w
 ill explore the challenges encountered and the performance levels, undersc
 oring SYCL's potential as an efficient GPU framework.\n\nDomain: Engineeri
 ng, Computational Methods and Applied Mathematics\n\nSession Chair: Szilár
 d Páll (KTH Royal Institute of Technology, PDC Center for High Performance
  Computing)
END:VEVENT
END:VCALENDAR