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:20210402T160544Z
LOCATION:Poster Module
DTSTART;TZID=America/New_York:20201119T083000
DTEND;TZID=America/New_York:20201119T170000
UID:submissions.supercomputing.org_SC20_sess337_rpost150@linklings.com
SUMMARY:Achieving the Performance of Global Adaptive Routing Using Local I
 nformation on Dragonfly through Deep Learning
DESCRIPTION:Posters, Research Posters\n\nAchieving the Performance of Glob
 al Adaptive Routing Using Local Information on Dragonfly through Deep Lear
 ning\n\nChaulagain, Liza, Chunduri, Yuan, Lang\n\nThe Universal Globally A
 daptive Load-balance Routing (UGAL) with global information, referred as U
 GAL-G, represents an ideal form of adaptive routing on Dragonfly. UGAL-G i
 s impractical to implement, however, since the global information cannot b
 e maintained accurately. Practical adaptive routing schemes, such as UGAL 
 with local information (UGAL-L), performs noticeably worse than UGAL-G. In
  this work, we investigate a machine learning approach for routing on Drag
 onfly. Specifically, we develop a machine learning-based routing scheme, c
 alled UGAL-ML,  that is capable of making routing decisions like UGAL-G ba
 sed only on the information local to each router. Our preliminary evaluati
 on indicates that UGAL-ML can achieve comparable performance to UGAL-G for
  some traffic patterns.\n\nRegistration Category: Tech Program Reg Pass, E
 xhibits Reg Pass
END:VEVENT
END:VCALENDAR