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:20211207T055407Z
LOCATION:Online
DTSTART;TZID=America/Chicago:20211115T171500
DTEND;TZID=America/Chicago:20211115T174000
UID:submissions.supercomputing.org_SC21_sess423_ws_mlhpce116@linklings.com
SUMMARY:Is Disaggregation Possible for HPC Cognitive Simulation?
DESCRIPTION:Workshop\n\nIs Disaggregation Possible for HPC Cognitive Simul
 ation?\n\nWyatt, Yamamoto, Tosi, Karlin, Van Essen\n\nCognitive simulation
  (CogSim) is an important and emerging workflow for HPC scientific explora
 tion and scientific machine learning (SciML).   One challenging workload f
 or CogSim is the replacement of one component in a complex physical simula
 tion with a fast, learned, surrogate model that is inside of the computati
 onal loop.  The execution of this in-the-loop inference is particularly ch
 allenging because it requires frequent inference across multiple possible 
 target models, can be on the simulation's critical path (latency bound), i
 s subject to requests from multiple MPI ranks, and typically contains a sm
 all number of samples per request.  In this paper we explore the use of la
 rge, dedicated Deep Learning / AI accelerators that are disaggregated from
  compute nodes for this CogSim workload.  We compare the trade-offs of usi
 ng these accelerators versus the node-local GPU accelerators on leadership
 -class HPC systems.\n\nTag: Online Only, Machine Learning and Artificial I
 ntelligence\n\nRegistration Category: Workshop Reg Pass
END:VEVENT
END:VCALENDAR