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:20230124T171519Z
LOCATION:C156
DTSTART;TZID=America/Chicago:20221114T093000
DTEND;TZID=America/Chicago:20221114T100000
UID:submissions.supercomputing.org_SC22_sess454_misc274@linklings.com
SUMMARY:Adding Malleability to MPI: Opportunities and Challenges
DESCRIPTION:Workshop\n\nAdding Malleability to MPI: Opportunities and Chal
 lenges\n\nSchulz\n\nThe Message Passing Interface (MPI) is the most domina
 nt programming model on HPC systems and has been instrumental in developin
 g efficient, large scale parallel applications. However, it has a rather s
 tatic view of compute resources building on top of the concept of immutabl
 e communicators. While this provides some easy-of-use and simplicity, it i
 s limiting, in particular for modern workflow-based workloads as well as i
 n its support for resource adaptive systems. The newly introduced concept 
 of MPI Sessions, however, opens the door more dynamicity and adaptivity. I
 n this talk I will highlight the opportunities that can arise from such di
 rections and discuss a novel approaches we are pursuing as part of several
  EuroHPC projects. Our ultimate goal is to provide full malleability in MP
 I as well as the surrounding software layers - from system software to app
 lications - and with that enable us to more efficiently harness the comput
 ational capabilities of current and future HPC systems.\n\nSession Format:
  Recorded\n\nTag: AI-HPC Convergence, Extreme Scale Computing, Parallel Pr
 ogramming Languages and Models, Performance, Runtime Systems\n\nRegistrati
 on Category: Workshop Reg Pass
END:VEVENT
END:VCALENDAR