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:20181221T160731Z
LOCATION:A2 Ballroom
DTSTART;TZID=America/Chicago:20181114T163000
DTEND;TZID=America/Chicago:20181114T170000
UID:submissions.supercomputing.org_SC18_sess466_gb105@linklings.com
SUMMARY:Exascale Deep Learning for Climate Analytics
DESCRIPTION:ACM Gordon Bell Finalist, Awards Presentation\n\n\nExascale De
 ep Learning for Climate Analytics\n\nKurth, Treichler, Romero, Mudigonda, 
 Luehr...\n\nWe extract pixel-level masks of extreme weather patterns using
  variants of Tiramisu and DeepLabv3+ neural networks. We describe improvem
 ents to the software frameworks, input pipeline, and the network training 
 algorithms necessary to efficiently scale deep learning on the Piz Daint a
 nd Summit systems. The Tiramisu network scales to 5300 P100 GPUs with a su
 stained throughput of 21.0 PF/s and parallel efficiency of 79.0%. DeepLabv
 3+ scales up to 27360 V100 GPUs with a sustained throughput of 325.8 PF/s 
 and a parallel efficiency of 90.7% in single precision. By taking advantag
 e of the FP16 Tensor Cores, a half-precision version of the DeepLabv3+ net
 work achieves a peak and sustained throughput of 1.13 EF/s and 999.0 PF/s 
 respectively.
URL:https://sc18.supercomputing.org/presentation/?id=gb105&sess=sess466
END:VEVENT
END:VCALENDAR