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:20260522T150120Z
LOCATION:C146
DTSTART;TZID=America/Chicago:20181114T103000
DTEND;TZID=America/Chicago:20181114T110000
UID:submissions.supercomputing.org_SC18_sess183_pap490@linklings.com
SUMMARY:Dynamic Tracing: Memoization of Task Graphs for Dynamic Task-Based
  Runtimes
DESCRIPTION:Wonchan Lee (Stanford University), Elliott Slaughter (SLAC Nat
 ional Accelerator Laboratory), Michael Bauer and Sean Treichler (Nvidia Co
 rporation), Todd Warszawski (Stanford University), Michael Garland (Nvidia
  Corporation), and Alex Aiken (Stanford University)\n\nMany recent program
 ming systems for both supercomputing and data center workloads generate ta
 sk graphs to express computations that run on parallel and distributed mac
 hines. Due to the overhead associated with constructing these graphs the d
 ependence analysis that generates them is often statically computed and me
 moized, and the resulting graph executed repeatedly at runtime. However, m
 any applications require a dynamic dependence analysis due to data depende
 nt behavior, but there are new challenges in capturing and re-executing ta
 sk graphs at runtime. In this work, we introduce dynamic tracing, a techni
 que to capture a dynamic dependence analysis of a trace that generates a t
 ask graph, and replay it. We show that an implementation of dynamic tracin
 g improves strong scaling by an average of 4.9X and up to 7.0X on a suite 
 of already optimized benchmarks.\n\nTag: Algorithms, Architectures, Memory
 , Networks, Parallel Programming Languages, Libraries, and Models, Power, 
 Programming Systems, Scheduling\n\nRegistration Category: Tech Program Reg
  Pass\n\nSession Chair: Patrick S. McCormick (Los Alamos National Laborato
 ry (LANL))\n\n
END:VEVENT
END:VCALENDAR