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:20260202T201806Z
LOCATION:261-262-265-266
DTSTART;TZID=America/Chicago:20251120T141500
DTEND;TZID=America/Chicago:20251120T143700
UID:submissions.supercomputing.org_SC25_sess162_pap770@linklings.com
SUMMARY:Stability-Preserving Lossy Compression for Large-Scale Partial Dif
 ferential Equations
DESCRIPTION:Qian Gong (Oak Ridge National Laboratory (ORNL)), Mark Ainswor
 th (Brown University), Jieyang Chen (Oregon University), Xin Liang (Univer
 sity of Kentucky), Liangji Zhu and Ethan Klasky (Florida University), Tush
 ar Athawale (Oak Ridge National Laboratory (ORNL)), Qing Liu (New Jersey I
 nstitute of Technology (CSLA)), Anand Rangarajan and Sanjay Ranka (Univers
 ity of Florida), and Scott Klasky (Oak Ridge National Laboratory (ORNL))\n
 \nCheckpoint/Restart (C/R) strategies are vital for fault tolerance in PDE
 -based scientific simulations, yet traditional checkpointing incurs signif
 icant I/O overhead. Lossy compression offers a scalable solution by reduci
 ng checkpoint data size, but conventional methods often lack control over 
 physical invariants (e.g., energy), leading to instability such as oscilla
 tions or divergence in partial differential equation (PDE) systems. This p
 aper introduces a stability-preserving compression approach tailored for P
 DE simulations by explicitly controlling kinetic and potential energy pert
 urbations to ensure stable restarts. Extensive experiments conducted acros
 s diverse PDE configurations demonstrate that our method maintains numeric
 al stability with minimal error magnification—even across multiple checkpo
 int-restart cycles—outperforming state-of-the-art lossy compressors. Paral
 lel evaluations on the Frontier supercomputer show up to 8.4× improvement 
 in checkpoint write performance and 6.3× in read performance, while mainta
 ining relative $L^2$ errors $\sim$2e-6 throughout continued simulation. Th
 ese results provide practical guidance for balancing compression accuracy,
  stability, and computational efficiency in large-scale PDE applications.\
 n\nTag: Algorithms, Applications, State of the Practice\n\nRecording: Live
 streamed, Recorded\n\nRegistration Category: Technical Program Reg Pass\n\
 nSession Chair: Shaikh Arifuzzaman (University of Nevada, Las Vegas)\n\n
END:VEVENT
END:VCALENDAR