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:20230124T171521Z
LOCATION:C1-2-3
DTSTART;TZID=America/Chicago:20221117T083000
DTEND;TZID=America/Chicago:20221117T170000
UID:submissions.supercomputing.org_SC22_sess226_spostg106@linklings.com
SUMMARY:SurrogateTrain: Drastically Improving Performance of Data Loading 
 for Training Scientific Surrogate Models
DESCRIPTION:ACM Student Research Competition: Graduate Poster, ACM Student
  Research Competition: Undergraduate Poster, Posters\n\nSurrogateTrain: Dr
 astically Improving Performance of Data Loading for Training Scientific Su
 rrogate Models\n\nSun\n\nDeep learning surrogate models have drawn much at
 tention in large-scale scientific simulations because they can provide sim
 ilar results to simulations at lower computational costs. To process large
  amounts of scientific data, distributed training on high-performance comp
 uting (HPC) clusters is often used. Training a surrogate model with data p
 arallelism consists of three major steps: (1) Each device loads a subset o
 f the dataset from the parallel filesystem; (2) Computing the model update
  on each device; (3) Communicating between devices to synchronize the mode
 l update. During these steps, we observe that data loading is the main per
 formance bottleneck for training surrogate models. To this end, we propose
  SurrogateTrain, an efficient data-loading approach for training surrogate
  models, including offline scheduling and on-demand buffering. Our evaluat
 ion on a scientific surrogate model demonstrates that SurrogateTrain reduc
 es the amount of data loaded by 6.7× and achieves up to 4.7× speedup in da
 ta loading.\n\nRegistration Category: Tech Program Reg Pass
END:VEVENT
END:VCALENDAR