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:20260422T000605Z
LOCATION:DEF Concourse
DTSTART;TZID=America/Denver:20231116T100000
DTEND;TZID=America/Denver:20231116T170000
UID:submissions.supercomputing.org_SC23_sess304_rpost125@linklings.com
SUMMARY:Scalable Reduced-Order Modeling for Three-Dimensional Turbulent Fl
 ow
DESCRIPTION:Kazuto Ando and Rahul Bale (RIKEN Center for Computational Sci
 ence (R-CCS); Kobe University, Japan); Akiyoshi Kuroda (RIKEN Center for C
 omputational Science (R-CCS)); and Makoto Tsubokura (RIKEN Center for Comp
 utational Science (R-CCS); Kobe University, Japan)\n\nA neural network-bas
 ed reduced order modeling method for three-dimensional turbulent flow simu
 lation is proposed. This method was implemented as the scalable distribute
 d learning on Fugaku. Our method constitutes a dimensional reduction using
  a convolutional-autoencoder-like neural network and the time evolution pr
 ediction using long short-term memory networks. The time evolution of the 
 turbulent three-dimensional flow (e.g., Re=2.8×10^6) could be simulated at
  a significantly lower cost (approximately four orders of magnitude) witho
 ut a major loss in accuracy. Using the single core memory group, our imple
 mentation shows 370 GFLOPS (24.28% of the peak performance) for the entire
  training loop and 753 GFLOPS (24.28% of the peak performance) for the con
 volution kernel. Our implementation scales up to 25,250 computational node
 s (1,212,000 cores). Thus it shows 72.9 % of weak scaling performance (7.8
  PFLOPS) for the entire training loop. On the other hand, the convolution 
 routine shows 100.8% of weak scaling performance (113 PFLOPS).\n\nTag: Art
 ificial Intelligence/Machine Learning, Architecture and Networks, Heteroge
 neous Computing, I/O and File Systems, Performance Measurement, Modeling, 
 and Tools, Post-Moore Computing, Programming Frameworks and System Softwar
 e, Quantum Computing\n\nRegistration Category: Tech Program Reg Pass, Exhi
 bits Reg Pass\n\n
END:VEVENT
END:VCALENDAR