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:20260522T150111Z
LOCATION:C2/3/4 Ballroom
DTSTART;TZID=America/Chicago:20181115T083000
DTEND;TZID=America/Chicago:20181115T170000
UID:submissions.supercomputing.org_SC18_sess324_post135@linklings.com
SUMMARY:Multi-GPU Accelerated Non-Hydrostatic Numerical Ocean Model with G
 PUDirect RDMA Transfers
DESCRIPTION:Takateru Yamagishi (Research Organization for Information Scie
 nce and Technology, Japan) and Yoshimasa Matsumura and Hiroyasu Hasumi (Un
 iversity of Tokyo)\n\nWe have implemented our “kinaco” numerical ocean mod
 el on Tokyo University’s Reedbush supercomputer, which utilizes the latest
  Nvidia Pascal P100 GPUs with GPUDirect technology. We have also optimized
  the model’s Poisson/Helmholtz solver by adjusting the global memory align
 ment and thread block configuration, introducing shuffle functions to acce
 lerate the creation of coarse grids and merging small kernels in the multi
 grid preconditioner. We also utilize GPUDirect RDMA transfers to improve M
 PI communication efficiency. By exploiting the GPUs’ capabilities, the GPU
  implementation is now twice as fast as the CPU version, and it shows good
  weak scalability to multiple GPUs. Most of the GPU kernels are accelerate
 d, and the velocity diagnosis functions in particular are now approximatel
 y seven times faster. The performance of inter-node data transfers using a
  CUDA-aware MPI library with GPUDirect RDMA transfers is comparable to tha
 t on CPUs.\n\nRegistration Category: Tech Program Reg Pass, Exhibits Reg P
 ass\n\n
END:VEVENT
END:VCALENDAR