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:20251118T105200
DTEND;TZID=America/Chicago:20251118T111500
UID:submissions.supercomputing.org_SC25_sess575_gb101@linklings.com
SUMMARY:Simulating Many-Engine Spacecraft: Exceeding 1 Quadrillion Degrees
  of Freedom via Information Geometric Regularization
DESCRIPTION:Benjamin Wilfong, Anand Radhakrishnan, Henry Le Berre, Daniel 
 Vickers, and Tanush Prathi (Georgia Institute of Technology); Nikolaos Tse
 lepidis and Benedikt Dorschner (NVIDIA); Reuben Budiardja (Oak Ridge Natio
 nal Laboratory); Brian Cornille (AMD); Stephen Abbott (Hewlett Packard Ent
 erprise (HPE)); Florian Schäfer (New York University); and Spencer Bryngel
 son (Georgia Institute of Technology)\n\nWe present an optimized implement
 ation of the recently proposed information geometric regularization (IGR) 
 for unprecedented scale simulation of compressible fluid flows applied to 
 multi-engine spacecraft boosters. We improve upon state-of-the-art computa
 tional fluid dynamics (CFD) techniques along computational cost, memory fo
 otprint, and energy-to-solution metrics.\n\nUnified memory on coupled CPU-
 GPU or APU platforms increases problem size with negligible overhead. Mixe
 d half/single-precision storage and computation on well-conditioned numeri
 cs is used. We simulate flow at 200 trillion grid points and 1 quadrillion
  degrees of freedom, exceeding the current record by a factor of 20. A fac
 tor of 4 wall-time speedup is achieved over optimized baselines. Ideal wea
 k scaling is seen on OLCF Frontier, LLNL El Capitan, and CSCS Alps using t
 he full systems. Strong scaling is near ideal at extreme conditions, inclu
 ding 80\% efficiency on CSCS Alps with an 8-node baseline and stretching t
 o the full system.\n\nTag: Applications, Gordon Bell Finalist\n\nRecording
 : Livestreamed, Recorded\n\nRegistration Category: Technical Program Reg P
 ass\n\nSession Chair: Amanda Randles (Duke University)\n\n
END:VEVENT
END:VCALENDAR