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:20210402T160557Z
LOCATION:Track 7
DTSTART;TZID=America/New_York:20201113T111500
DTEND;TZID=America/New_York:20201113T114500
UID:submissions.supercomputing.org_SC20_sess226_ws_rosscur102@linklings.co
 m
SUMMARY:A Dataflow-Graph Partitioning Method for Training Large Deep Learn
 ing Models
DESCRIPTION:Workshop\n\nA Dataflow-Graph Partitioning Method for Training 
 Large Deep Learning Models\n\nUnat, Qararyah\n\nLarge Deep Neural Network 
 (DNNs) models have substantial memory requirements to store the model para
 meters and intermediate results. As a result, the limited device memory be
 comes a bottleneck when training those models. We propose a deterministic,
  generic and efficient partitioning strategy for DNNs that are represented
  as computational graphs. The proposed partitioning algorithm decides a pl
 acement of a DNN’s underlying computational graph operations across multip
 le accelerators, so that the memory constraints of the devices are met and
  the training time is minimized. To the best of our knowledge, the strateg
 y deployed in this work is the first that has absolute independence of the
  structure and operation types in DNN models. Therefore, it guarantees fut
 ure compatibility and can be used with any type of emerging model, even if
  it has zero resemblance to the existing models in terms of structure or e
 ven the nature of the learning process and its operations. In this talk, I
  will be presenting the details of the method along with some performance 
 data and comparison with related work.\n\nTag: System Software and Runtime
  Systems\n\nRegistration Category: Workshop Reg Pass
END:VEVENT
END:VCALENDAR