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:20260202T201804Z
LOCATION:276
DTSTART;TZID=America/Chicago:20251117T143500
DTEND;TZID=America/Chicago:20251117T144000
UID:submissions.supercomputing.org_SC25_sess195_ws_whpc119@linklings.com
SUMMARY:Overhead Quantification of the Lightweight Distributed Metric Serv
 ice for High-Performance Computers
DESCRIPTION:Alex Knigge, M. Scot Swan, and Jennifer Green (Sandia National
  Laboratories)\n\nThe Lightweight Distributed Metric Service (LDMS) is a m
 onitoring framework that collects high-fidelity, high-volume node-level da
 ta on large distributed computer systems. LDMS is built to introduce negli
 gible overhead in application workloads which has been verified in several
  scale tests since its inception in 2014. However, new communication strat
 egies, sensor samplers, and fundamental data structures within the core LD
 MS code have been introduced that could increase the overhead. In this stu
 dy, we quantify the current overhead that LDMS introduces and verify that 
 it is insignificant. This was done through a variety of benchmarks and app
 lications where we captured timing and performance statistics while LDMS r
 an with different configurations.\n\nRecording: Partially Livestreamed, Pa
 rtially Recorded\n\nRegistration Category: Technical Program Reg Pass, Wor
 kshop Reg Pass\n\nSession Chairs: Jessica Imlau Dagostini (University of C
 alifornia, Santa Cruz); Elsa J. Gonsiorowski (Lawrence Livermore National 
 Laboratory (LLNL)); and Mozhgan Kabiri chimeh (NVIDIA Corporation)\n\n
END:VEVENT
END:VCALENDAR