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:505
DTSTART;TZID=America/Denver:20231113T142100
DTEND;TZID=America/Denver:20231113T142400
UID:submissions.supercomputing.org_SC23_sess442_ws_whpc101@linklings.com
SUMMARY:Exploring the Potential of GPU-initiated Communications in HPC App
 lications
DESCRIPTION:Nan Ding (Lawrence Berkeley National Laboratory (LBNL))\n\nThe
  two-sided MPI has become a de facto standard for communication on a distr
 ibuted memory system. As high-performance GPU computing becomes the trend,
  some numerical methods have a relatively simple communication pattern adh
 ering to the BSP model find MPI and its CUDA-aware variant can satisfy the
  performance requirements. Conversely, DAG-like computations have a more c
 omplex communication pattern and find it hard to scale to multiple GPUs. T
 hus, GPU-initiated one-sided communication becomes a viable solution for m
 ulti-GPU scaling. However, the lack of deep understanding of GPU-initiated
  one-sided communication and the real impact on applications performance b
 ecome a hurdle. In this work, we use a multi-GPU SpTRSV, which is used in 
 conjunction with Sparse LU for solving sparse linear systems, either as a 
 direct solver or as a precondition, to demonstrate our multi-GPU SpTRSV im
 plementation using NVSHMEM achieves up to 3x speedup on Perlmutter compare
 d to a single GPU implementation.\n\nTag: State of the Practice\n\nRegistr
 ation Category: Workshop Reg Pass\n\nSession Chairs: Elsa J. Gonsiorowski 
 (Lawrence Livermore National Laboratory (LLNL)) and Mozhgan Kabiri chimeh 
 (NVIDIA Corporation)\n\n
END:VEVENT
END:VCALENDAR