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:20210402T160556Z
LOCATION:Track 11
DTSTART;TZID=America/New_York:20201112T163000
DTEND;TZID=America/New_York:20201112T170000
UID:submissions.supercomputing.org_SC20_sess219_ws_prot101@linklings.com
SUMMARY:Empirical Modeling of Spatially Diverging Performance
DESCRIPTION:Workshop\n\nEmpirical Modeling of Spatially Diverging Performa
 nce\n\nCalotoiu, Geisenhofer, Kummer, Ritter, Weber...\n\nA common simplif
 ication made when modeling the performance of a parallel program is the as
 sumption that the performance behavior of all processes or threads is larg
 ely uniform. Empirical performance-modeling tools such as Extra-P exploit 
 this common pattern to make their modeling process more noise resilient, m
 itigating the effect of outliers by summarizing performance measurements o
 f individual functions across all processes. While the underlying assumpti
 on does not hold equally for all applications, knowing the qualitative dif
 ferences in how the performance of individual processes changes as executi
 on parameters are varied can reveal important performance bottlenecks such
  as malicious patterns of load imbalance. A challenge for empirical modeli
 ng tools, however, arises from the fact that the behavioral class of a pro
 cess may depend on the process configuration, letting process ranks migrat
 e between classes as the number of processes grows. In this paper, we intr
 oduce a novel approach to the problem of modeling of spatially diverging p
 erformance based on a certain type of process clustering. We apply our tec
 hnique to identify a previously unknown performance bottleneck in the BoSS
 S fluid-dynamics code. Removing it made the code regions in question run u
 p to 20x and the application as a whole run up to 4.5x faster.\n\nRegistra
 tion Category: Workshop Reg Pass
END:VEVENT
END:VCALENDAR