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:20261107T045506Z
LOCATION:W181c
DTSTART;TZID=America/Chicago:20261115T114200
DTEND;TZID=America/Chicago:20261115T120600
UID:submissions.supercomputing.org_SC26_sess110_ws_p3hpc125@linklings.com
SUMMARY:A Performance-Portable GPU-Accelerated Solution for Particle-Parti
 cle Particle-Mesh Method using Cabana and FFTX
DESCRIPTION:Het Mankad (Oak Ridge National Laboratory (ORNL)), Zoe Barbeau
  (Stanford University), Phillip Colella (Lawrence Berkeley National Labora
 tory), Sanil Rao and Franz Franchetti (Carnegie Mellon University), and Je
 ffrey Vetter and Keita Teranishi (Oak Ridge National Laboratory)\n\nOne ap
 proach to the development of high-performance computing applications is ba
 sed on libraries that implement algorithmic motifs in a portable, performa
 nt and concise fashion. We develop a GPU-accelerated algorithm for Method 
 of Local Corrections, a particle-particle particle-mesh method for simulat
 ing incompressible flow in 3D, using such motif libraries. The algorithm c
 omprises of structured grid calculations, local particle interactions, and
  free-space convolution. We use two motif libraries: Cabana, a motif libra
 ry for particles, built on top of Kokkos; and FFTX for fast Fourier Transf
 orm, that uses the SPIRAL code generation system. By using motif libraries
 , we obtain a concise and portable implementation (2K lines of code). We c
 ompare the measured performance to that predicted by the roofline model an
 d obtain 52\%-86\% on NVIDIA A100, 42\%-61\% on NVIDIA H100, 22\%-42\% on 
 AMD MI100, and 17\%-38\% on AMD MI300A.\n\nTag: Profiling and Optimization
  Tools, Programming Frameworks\n\nRecording: Livestreamed, Recorded\n\nReg
 istration Category: Technical Program Reg Pass, Workshop Reg Pass\n\nSessi
 on Chairs: CJ Newburn (NVIDIA Corporation), Scott J. Parker (Argonne Natio
 nal Laboratory), John Pennycook (Intel Corporation), and Kenneth Weiss (La
 wrence Livermore National Laboratory (LLNL))\n\n
END:VEVENT
END:VCALENDAR
