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:20211114T113000
DTEND;TZID=America/Chicago:20211114T120000
UID:submissions.supercomputing.org_SC21_sess430_ws_exampi103@linklings.com
SUMMARY:Overlapping Communication and Computation with ExaMPI's Strong Pro
 gress and Modern C++ Design
DESCRIPTION:Workshop\n\nOverlapping Communication and Computation with Exa
 MPI's Strong Progress and Modern C++ Design\n\nSchafer, Hines, Suggs, Rüfe
 nacht, Skjellum\n\nExaMPI is a modern, C++17+ MPI implementation designed 
 for modularity, extensibility, and understandability. In this work, we ove
 rview  functionality new to ExaMPI since its initial release, including Li
 bfabric-based network transport support. We also explain our rationale for
  why and how we choose to add new MPI features (and defer others). We also
  measured the latency of the aforementioned transports in ExaMPI and found
  that ExaMPI, while having slightly higher latency than other production M
 PIs, is competitive.  Lastly, we tested the ability for ExaMPI's strong pr
 ogress engine to enable overlap communication and computation, finding tha
 t considerable overlap is achieved without needing additional MPI helper c
 alls such as MPI_Test.\n\nTag: Online Only, Extreme Scale Computing, Paral
 lel Programming Languages and Models, Performance\n\nRegistration Category
 : Workshop Reg Pass
END:VEVENT
END:VCALENDAR