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:20260522T150118Z
LOCATION:D165
DTSTART;TZID=America/Chicago:20181112T103000
DTEND;TZID=America/Chicago:20181112T110000
UID:submissions.supercomputing.org_SC18_sess161_ws_pmbsf106@linklings.com
SUMMARY:Deep Learning at Scale on Nvidia V100 Accelerators
DESCRIPTION:Rengan Xu, Frank Han, and Quy Ta (Dell EMC)\n\nThe recent expl
 osion in the popularity of Deep Learning (DL) is due to a combination of i
 mproved algorithms, access to large datasets and increased computational p
 ower. This had led to a plethora of open-source DL frameworks, each with v
 arying characteristics and capabilities. End users are then left with the 
 difficult task of determining software and hardware configurations to get op
 timal performance from each framework. \n\nWe share our experiences and de
 velop best practices for DL training with TensorFlow, MXNet, and Caffe2. T
 he paper also looks at DL inferencing with TensorRT on Nvidia V100 “Volta”
  GPUs. It focuses on one of the more prominent neural network architecture
 s, Resnet50, combined with Imagenet dataset. We quantify the impact of har
 dware attributes on DL workloads such as the usage of PCIe vs NVLink GPUs,
  performance past a single worker node, effect of high speed interconnect 
 such as InfiniBand EDR on training, and the implication of utilizing a netw
 ork attached storage and its advantages.\n\nTag: Benchmarks, Parallel Prog
 ramming Languages, Libraries, and Models, Performance, Simulation\n\nRegis
 tration Category: Workshop Reg Pass\n\nSession Chair: Steven A. Wright (Un
 iversity of York, England)\n\n
END:VEVENT
END:VCALENDAR