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:20260202T201803Z
LOCATION:242
DTSTART;TZID=America/Chicago:20251116T103500
DTEND;TZID=America/Chicago:20251116T104500
UID:submissions.supercomputing.org_SC25_sess216_ws_isav107@linklings.com
SUMMARY:Lightning Talk: Predicting Optimal Rendering Frequency for In Situ
  Visualization in GPU–CPU Workflows
DESCRIPTION:Colleen Heinemann (University of Illinois at Urbana-Champaign)
 ; Jens Henrik Göbbert and Mathis Bode (Forschungzentrum Juelich, Jülich Su
 percomputing Centre (JSC)); Mengjiao Han, Joseph A. Insley, Silvio Rizzi, 
 and Janet Knowles (Argonne National Laboratory (ANL)); Michael E. Papka (A
 rgonne National Laboratory (ANL), University of Illinois Chicago); and Vic
 tor A. Mateevitsi (Argonne National Laboratory (ANL))\n\nHigh Performance 
 Computing (HPC) simulations often run on GPUs while in situ rendering task
 s are offloaded to CPUs. Deciding how frequently to perform these in situ 
 renderings is a challenge: render too frequently, and the simulation, or p
 roducer, oversaturates the rendering pipeline, or consumer; render too spa
 rsely, and the CPU resources remain idle. This project seeks to develop ma
 chine learning (ML) models that can predict rendering times based on avail
 able system resources as well as simulation parameters. By leveraging ML-d
 riven insight, the goal of this project is to analyze the tradeoffs and de
 termine an optimal rendering interval for simulations such as nekRS instru
 mented with Ascent, in this instance. and ensure balanced workloads betwee
 n the simulation and rendering tasks, ultimately improving overall computa
 tional efficiency.\n\nTag: AI, Machine Learning, & Deep Learning, Clouds &
  Distributed Computing, Performance Evaluation, Scalability, & Portability
 , Scientific & Information Visualization\n\nRecording: Partially Livestrea
 med, Partially Recorded\n\nRegistration Category: Technical Program Reg Pa
 ss, Workshop Reg Pass\n\nSession Chairs: Axel Huebl (Lawrence Berkeley Nat
 ional Laboratory (LBNL)); Will Usher (Luminary Cloud Inc); Nicola Ferrier 
 (Argonne National Laboratory (ANL), University of Chicago); Estelle Dirand
  (TotalEnergies, France); Victor Mateevitsi (Argonne National Laboratory (
 ANL), University of Illinois Chicago); and Jens Henrik Göbbert (Forschungs
 zentrum Jülich, Jülich Supercomputing Centre (JSC))\n\n
END:VEVENT
END:VCALENDAR