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:20211207T055342Z
LOCATION:227
DTSTART;TZID=America/Chicago:20211115T110000
DTEND;TZID=America/Chicago:20211115T113000
UID:submissions.supercomputing.org_SC21_sess347_ws_scsc104@linklings.com
SUMMARY:Checkpoint-Restart Libraries Must Become More Fault Tolerant
DESCRIPTION:Workshop\n\nCheckpoint-Restart Libraries Must Become More Faul
 t Tolerant\n\nSkjellum, Schafer\n\nProduction MPI codes need checkpoint-re
 start (CPR) support. Clearly, checkpoint-restart libraries  must be fault 
 tolerant lest they open up a window of vulnerability for failures  with by
 zantine outcomes. But, certain popular libraries that leverage MPI  are ev
 idently not fault tolerant.\nNowadays, fault detection with automatic reco
 very without batch requeueing is a strong requirement for production envir
 onments. Thus, allowing deadlock and setting long timeouts  are suboptimal
  for fault detection even when paired with conservative recovery from the 
 penultimate checkpoint. \n\nWhen MPI is used as a communication mechanism 
 within a CPR library, such libraries must offer fault-tolerant extensions 
 with minimal detection, isolation, mitigation, and potential recovery sema
 ntics to aid the CPR's library fail-backward.  Communication between MPI a
 nd the checkpoint library regarding system health may be valuable.  For fa
 ult-tolerant MPI programs (e.g.,  using APIs like FA-MPI,  Stages/Reinit, 
 or ULFM), the checkpoint library must cooperate with the extended model or
  else invalidate  fault-tolerant operation.\n\nTag: Parallel Programming L
 anguages and Models, Reliability and Resiliency\n\nRegistration Category: 
 Workshop Reg Pass
END:VEVENT
END:VCALENDAR