BEGIN:VCALENDAR
VERSION:2.0
PRODID:Linklings LLC
BEGIN:VTIMEZONE
TZID:America/New_York
X-LIC-LOCATION:America/New_York
BEGIN:DAYLIGHT
TZOFFSETFROM:-0500
TZOFFSETTO:-0400
TZNAME:EDT
DTSTART:19700308T020000
RRULE:FREQ=YEARLY;BYMONTH=3;BYDAY=2SU
END:DAYLIGHT
BEGIN:STANDARD
TZOFFSETFROM:-0400
TZOFFSETTO:-0500
TZNAME:EST
DTSTART:19701101T020000
RRULE:FREQ=YEARLY;BYMONTH=11;BYDAY=1SU
END:STANDARD
END:VTIMEZONE
BEGIN:VEVENT
DTSTAMP:20250626T233526Z
LOCATION:B302-B305
DTSTART;TZID=America/New_York:20241120T100000
DTEND;TZID=America/New_York:20241120T170000
UID:submissions.supercomputing.org_SC24_sess533_post160@linklings.com
SUMMARY:Identifying Regions of Non-Determinism in HPC Simulations Through 
 Event Graph Alignment
DESCRIPTION:Dhroov Pandey (University of North Texas)\n\nHigh performance 
 computing (HPC) applications using MPI (Message Passing Interface) often f
 ace non-determinism (ND) due to asynchronous MPI calls, making ND source i
 dentification challenging. Modeling execution as an event graph, where MPI
  calls are nodes and communication is edges, can be useful. Focusing on Me
 ssage ND, which involves variability in MPI communication order across run
 s, we detect potential ND sources by comparing edge sets between event gra
 phs. Accurate comparison requires aligning event graph nodes, but traditio
 nal methods like NetAlign, graphlet degree vectors, and Graph Auto-Encoder
 s struggle due to the regularity of event graphs. We propose a meta graph 
 heuristic utilizing structural constraints and a message passing scheme fo
 r sparse directed acyclic graphs, achieving up to 70% improvement in align
 ment accuracy over conventional techniques.\n\nRegistration Category: Tech
  Program Reg Pass, Exhibits Reg Pass\n\nSession Chairs: Ayesha Afzal (Frie
 drich-Alexander University, Erlangen-Nuremberg; Erlangen National High Per
 formance Computing Center); Sally Ellingson (University of Kentucky); and 
 Alan Sussman (University of Maryland)\n\n
END:VEVENT
END:VCALENDAR