BEGIN:VCALENDAR
VERSION:2.0
PRODID:Linklings LLC
BEGIN:VTIMEZONE
TZID:America/New_York
X-LIC-LOCATION:America/New_York
BEGIN:DAYLIGHT
TZOFFSETFROM:-0500
TZOFFSETTO:-0400
TZNAME:EDT
DTSTART:19700308T020000
RRULE:FREQ=YEARLY;BYMONTH=3;BYDAY=2SU
END:DAYLIGHT
BEGIN:STANDARD
TZOFFSETFROM:-0400
TZOFFSETTO:-0500
TZNAME:EST
DTSTART:19701101T020000
RRULE:FREQ=YEARLY;BYMONTH=11;BYDAY=1SU
END:STANDARD
END:VTIMEZONE
BEGIN:VEVENT
DTSTAMP:20210402T160555Z
LOCATION:Track 8
DTSTART;TZID=America/New_York:20201112T170000
DTEND;TZID=America/New_York:20201112T172500
UID:submissions.supercomputing.org_SC20_sess214_ws_lasalss114@linklings.co
m
SUMMARY:A Fast Scalable Iterative Implicit Solver with Green's Function-Ba
sed Neural Networks
DESCRIPTION:Workshop\n\nA Fast Scalable Iterative Implicit Solver with Gre
en's Function-Based Neural Networks\n\nIchimura, Fujita, Hori, Maddegedara
, Ueda...\n\nBased on the Green's functions that reflect mathematical prop
erties of partial differential equations (PDE), we developed a novel preco
nditioner using neural networks (NNs) with high accuracy and small computa
tional cost for improving the convergence property of an iterative implici
t solver. As the dense and uniform computation involved in NNs are more ef
ficient than that of the conventional PDE solver schemes, we could solve t
he time evolution of a 405,017,091 degrees-of-freedom highly heterogeneous
problem in 5.48-fold shorter time compared to a typical PDE solver. The m
ethod is also suitable for use with low-precision arithmetic in NNs as the
accuracy of the final solution is guaranteed. The localized property of N
Ns enable high scalability for solving large problems (103,305,758,211 deg
rees-of-freedom problem solved with 97.4% weak scalability using 256 Casca
de Lake Xeon CPU-based Oakbridge-CX nodes with a total of 14336 CPU cores
with developed MPI-OpenMP hybrid code). This method can be used in various
PDE-based simulations and has potential to make broad ripple effects in v
arious fields.\n\nTag: Algorithms, Extreme Scale Computing, Performance/Pr
oductivity Measurement and Evaluation, Scalable Computing, Scientific Comp
uting\n\nRegistration Category: Workshop Reg Pass
END:VEVENT
END:VCALENDAR