BEGIN:VCALENDAR
VERSION:2.0
PRODID:Linklings LLC
BEGIN:VTIMEZONE
TZID:America/Denver
X-LIC-LOCATION:America/Denver
BEGIN:DAYLIGHT
TZOFFSETFROM:-0700
TZOFFSETTO:-0600
TZNAME:MDT
DTSTART:19700308T020000
RRULE:FREQ=YEARLY;BYMONTH=3;BYDAY=2SU
END:DAYLIGHT
BEGIN:STANDARD
TZOFFSETFROM:-0600
TZOFFSETTO:-0700
TZNAME:MST
DTSTART:19701101T020000
RRULE:FREQ=YEARLY;BYMONTH=11;BYDAY=1SU
END:STANDARD
END:VTIMEZONE
BEGIN:VEVENT
DTSTAMP:20260422T000713Z
LOCATION:505
DTSTART;TZID=America/Denver:20231112T160000
DTEND;TZID=America/Denver:20231112T162000
UID:submissions.supercomputing.org_SC23_sess428_ws_ipdrm101@linklings.com
SUMMARY:Dask-Extended External Tasks for HPC/ML In Transit Workflows
DESCRIPTION:Amal Gueroudji (Argonne National Laboratory (ANL)), Julien Big
 ot (Atomic Energy and Alternative Energies Commission (CEA)), Bruno Raffin
  (INRIA), and Robert Ross (Argonne National Laboratory (ANL))\n\nIn situ w
 orkflows are inescapable to fully leverage exascale architectures. They ca
 n however be complex to build as simulation and data analytics come from t
 wo different software ecosystems with their own paradigms. We extend deisa
  by introducing the concept of external tasks to support the description o
 f analytics graphs spanning multiple timesteps ahead of time while improvi
 ng scalability. This new approach leads to straightforward support for con
 tracts to limit the data transferred to that actually analyzed in a given 
 execution. We implement this approach using Dask and MPI and evaluate it u
 sing an in-transit workflow that uses an unsupervised ML model. We compare
  our work to plain Dask and to the previous version of deisa. Our work per
 forms better, up to ×7, for the simulation and ×3, for the analytics compa
 red to deisa and ×18 less costly compared to plain Dask. All of those with
  similar development efforts.\n\nTag: Distributed Computing, Middleware an
 d System Software, Runtime Systems\n\nRegistration Category: Workshop Reg 
 Pass\n\nSession Chairs: Barbara Chapman (Hewlett Packard Enterprise (HPE),
  Stony Brook University); Joseph Manzano (Pacific Northwest National Labor
 atory (PNNL)); Shirley Moore (University of Texas at El Paso); EunJung (EJ
 ) Park (Qualcomm); and Joshua Suetterlein (Pacific Northwest National Labo
 ratory (PNNL))\n\n
END:VEVENT
END:VCALENDAR