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:20230124T171523Z
LOCATION:C147-154
DTSTART;TZID=America/Chicago:20221113T140000
DTEND;TZID=America/Chicago:20221113T143000
UID:submissions.supercomputing.org_SC22_sess427_ws_exampi103@linklings.com
SUMMARY:IPMPI: Improved MPI Communication Logger
DESCRIPTION:Workshop\n\nIPMPI: Improved MPI Communication Logger\n\nAgrawa
 l, Malakar\n\nMost parallel profilers reveal only the MPI point-to-point c
 ommunication patterns in a communication matrix. However, this may lead to
  sub-optimal process mapping due to incomplete information about the appli
 cation communications. In this work, we have developed a profiler, IPMPI, 
 that reveals the complete communication information including collective c
 ommunications. We compare the results of two recent process mapping tools,
  TopoMatch and LPMS, with and without the input from IPMPI. We have run si
 x applications and benchmarks such as ROMS, LAMMPS, DFT, HPCG, miniAMR and
  miniFE on two supercomputers and our shared department cluster. We observ
 ed a maximum average gain between 5.7 - 8.6% across these systems.\n\nSess
 ion Format: Recorded\n\nRegistration Category: Workshop Reg Pass
END:VEVENT
END:VCALENDAR