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:20211207T055404Z
LOCATION:Online
DTSTART;TZID=America/Chicago:20211114T160000
DTEND;TZID=America/Chicago:20211114T163000
UID:submissions.supercomputing.org_SC21_sess430_ws_exampi104@linklings.com
SUMMARY:A Benchmark to Understand Communication Performance in Hybrid MPI 
 and GPU Applications
DESCRIPTION:Workshop\n\nA Benchmark to Understand Communication Performanc
 e in Hybrid MPI and GPU Applications\n\nHaskins, Bridges, Levy, Ferreira\n
 \nAnalyzing MPI communication costs on extreme-scale HPC systems is critic
 al to ensuring optimal performance. Factors including scalability and wide
 spread use of GPUs complicate this analysis.  To address this challenge, w
 e need benchmarks and tools that use GPU and host memory in a manner simil
 ar to production applications.  These tools need to provide sufficient opt
 ions to, for example, fine-tune and analyze the costs of low-level data ex
 change between host and device memory. In this extended abstract, we descr
 ibe: the prototype implementation of a GPU-focused communication benchmark
  written in Kokkos/C++ based on FIESTA; and an MPI profiling library integ
 rated with NVIDIA’s nvprof profiling tool.  Using this combination, we dem
 onstrate how GPU data movement overheads can vary and examine how these to
 ols help us understand performance behavior.  These tools are important co
 ntributions that we believe will guide data movement decisions on extreme-
 scale HPC systems.\n\nTag: Online Only, Extreme Scale Computing, Parallel 
 Programming Languages and Models, Performance\n\nRegistration Category: Wo
 rkshop Reg Pass
END:VEVENT
END:VCALENDAR