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:20260202T201805Z
LOCATION:241
DTSTART;TZID=America/Chicago:20251117T163000
DTEND;TZID=America/Chicago:20251117T170000
UID:submissions.supercomputing.org_SC25_sess225_ws_prot106@linklings.com
SUMMARY:Scalable, High-Fidelity Monitoring of Application Communication Pa
 tterns in Vernier
DESCRIPTION:Jered Dominguez-Trujillo (University of New Mexico, Los Alamos
  National Laboratory (LANL)); Derek Schafer (University of New Mexico); Ri
 ley Shipley (Tennessee Tech University); Ryan Marshall (Los Alamos Nationa
 l Laboratory (LANL)); Nicholas Bacon (University of New Mexico); Maxim Mor
 aru and Galen Shipman (Los Alamos National Laboratory (LANL)); Anthony Skj
 ellum (Tennessee Tech University); and Patrick Bridges (University of New 
 Mexico)\n\nUnderstanding the irregular, dynamic communication patterns in 
 HPC applications at scale is critical when evaluating potential software o
 ptimizations and hardware architectures. Current systems monitor communica
 tion behavior for entire applications as exhaustive traces or general-purp
 ose aggregated statistics. Generally, these approaches often do not scale 
 well and the data gathered is often too generic or inflexible to make spec
 ific hardware/software optimizations. This paper describes a new, configur
 able, histogram-based approach to gathering scalable, high-fidelity monito
 ring information about HPC communication that we implemented in the Vernie
 r communication monitoring system. This approach enables targeted collecti
 on of statistical data about annotated communication patterns for online o
 r offline analysis, benchmarking, or network simulations. We assess these 
 capabilities by collecting communication patterns from several production 
 HPC applications at scale, showing that the resulting statistical represen
 tations accurately characterize the communication patterns in these applic
 ations, and can be used to provide new insights into communication pattern
 s of complex HPC applications.\n\nRecording: Livestreamed, Recorded\n\nReg
 istration Category: Technical Program Reg Pass, Workshop Reg Pass\n\nSessi
 on Chairs: David Boehme (Lawrence Livermore National Laboratory (LLNL)); A
 nthony Danalis (AMD); and Josef Weidendorfer (Leibniz Supercomputing Centr
 e (LRZ), Technical University of Munich)\n\n
END:VEVENT
END:VCALENDAR