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:20181221T160727Z
LOCATION:D171
DTSTART;TZID=America/Chicago:20181111T155500
DTEND;TZID=America/Chicago:20181111T162000
UID:submissions.supercomputing.org_SC18_sess152_ws_vpa104@linklings.com
SUMMARY:Using Deep Learning for Automated Communication Pattern Characteri
 zation: Little Steps and Big Challenges
DESCRIPTION:Workshop\nData Analytics, Performance, Visualization, Workshop
  Reg Pass\n\nUsing Deep Learning for Automated Communication Pattern Chara
 cterization: Little Steps and Big Challenges\n\nRoth, Huck, Gopalakrishnan
 , Wolf\n\nCharacterization of a parallel application's communication patte
 rns can be useful for performance analysis, debugging, and system design. 
 However, obtaining and interpreting a characterization can be difficult. A
 Chax implements an approach that uses search and a library of known commun
 ication patterns to automatically characterize communication patterns. Our
  approach has some limitations that reduce its effectiveness for the patte
 rns and pattern combinations used by some real-world applications. By view
 ing AChax's pattern recognition problem as an image recognition problem, i
 t may be possible to use deep learning to address these limitations. In th
 is position paper, we present our current ideas regarding the benefits and
  challenges of integrating deep learning into AChax and our conclusion tha
 t a hybrid approach combining deep learning classification, regression, an
 d the existing AChax approach may be the best long-term solution to the pr
 oblem of parameterizing recognized communication patterns.
URL:https://sc18.supercomputing.org/presentation/?id=ws_vpa104&sess=sess15
 2
END:VEVENT
END:VCALENDAR

