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:20260202T201249Z
LOCATION:Second Floor Atrium
DTSTART;TZID=America/Chicago:20251120T080000
DTEND;TZID=America/Chicago:20251120T170000
UID:submissions.supercomputing.org_SC25_sess533_post292@linklings.com
SUMMARY:Explicit Low-Order Finite-Element Wave Simulation Accelerated with
  Variable-Precision Computing Using INT8 Tensor Cores
DESCRIPTION:Kohei Fujita and Tsuyoshi Ichimura (The University of Tokyo, R
 IKEN); Muneo Hori (Japan Agency for Marine-Earth Science and Technology); 
 and Lalith Maddegedara (The University of Tokyo)\n\nUsing low-precision co
 res for acceleration of PDE-based simulations with sparse or small matrice
 s is often challenging due to the frequent data conversion between high- a
 nd low-precision variables, and that the required precision varies in time
 /space due to the heterogeneity of the target problem. As an example of ac
 celerating such PDE-based simulations, we develop an integer-based variabl
 e-precision computing method with low data-conversion costs for low-order 
 explicit finite-element wave propagation simulations. Here, the precision 
 level used for solving the problem is chosen locally to attain simulation 
 accuracy, and is accelerated using INT8 Tensor Cores. This leads to a 3.3-
 fold speedup from a baseline FP64 CUDA-core-based implementation with equi
 valent simulation accuracy, with 87% weak efficiency up to 256 compute nod
 es of the GH200-based Miyabi supercomputer. These ideas are expected to be
  useful for accelerating other PDE-based problems with sparse or small mat
 rices on computer architectures with high-performance, low-precision cores
 .\n\nTag: Research & ACM SRC Posters\n\nRegistration Category: Technical P
 rogram Reg Pass\n\nSession Chairs: Kento Sato (RIKEN Center for Computatio
 nal Science (R-CCS)); Chris Schlipalius (Pawsey Supercomputing Research Ce
 ntre; Commonwealth Scientific and Industrial Research Organisation (CSIRO)
 , Australia); and Anja Gerbes (Georg-August-Universität Göttingen)\n\n
END:VEVENT
END:VCALENDAR