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:20210402T160552Z
LOCATION:Track 4
DTSTART;TZID=America/New_York:20201113T123000
DTEND;TZID=America/New_York:20201113T125000
UID:submissions.supercomputing.org_SC20_sess223_ws_hpcsysp106@linklings.co
 m
SUMMARY:Log-Based Identification, Classification, and Behavior Prediction 
 of HPC Applications
DESCRIPTION:Workshop\n\nLog-Based Identification, Classification, and Beha
 vior Prediction of HPC Applications\n\nLewis, Liu, Kettimuthu, Papka\n\nLe
 adership supercomputers, such as those operated by the Argonne Leadership 
 Computing Facility (ALCF), provide an important avenue for scientific expl
 oration and discovery, enabling simulation, data analysis and visualizatio
 n, and artificial intelligence at massive scale. As we move into the exasc
 ale supercomputing era in 2021 with the advent of Aurora, Frontier, and ot
 her exascale machines, it's important that we are able to understand the i
 nteractions between the applications being run, and the hardware they run 
 on, to optimize the use of these expensive and high-demand resources. <br 
 /><br />In previous work, we analyzed a collection of production machine s
 cheduling and performance logs to better understand application behaviors 
 and characteristics. This work further refines our understanding of how sc
 ientific users leverage leadership computing resources; we show that syste
 m-level hardware performance counters can work as a lightweight, low-overh
 ead alternative to more performance-intensive benchmarking and logging ins
 trumentation for certain data analysis tasks.  We also demonstrate a 
 method for predicting application runtimes on leadership computing resourc
 es using data gathered from logging sources at submission.\n\nRegistration
  Category: Workshop Reg Pass
END:VEVENT
END:VCALENDAR