BEGIN:VCALENDAR
VERSION:2.0
PRODID:Linklings LLC
BEGIN:VTIMEZONE
TZID:Europe/Stockholm
X-LIC-LOCATION:Europe/Stockholm
BEGIN:DAYLIGHT
TZOFFSETFROM:+0100
TZOFFSETTO:+0200
TZNAME:CEST
DTSTART:19700308T020000
RRULE:FREQ=YEARLY;BYMONTH=3;BYDAY=-1SU
END:DAYLIGHT
BEGIN:STANDARD
TZOFFSETFROM:+0200
TZOFFSETTO:+0100
TZNAME:CET
DTSTART:19701101T020000
RRULE:FREQ=YEARLY;BYMONTH=10;BYDAY=-1SU
END:STANDARD
END:VTIMEZONE
BEGIN:VEVENT
DTSTAMP:20260522T162632Z
LOCATION:Bldg. 6 - Room 104
DTSTART;TZID=Europe/Stockholm:20260629T113000
DTEND;TZID=Europe/Stockholm:20260629T120000
UID:submissions.pasc-conference.org_PASC26_sess171_pap133@linklings.com
SUMMARY:Solver-Integrated Lossy and Lossless Compression for Scalable Flow
  Simulations
DESCRIPTION:Viral Sudip Shah (University of Illinois Urbana-Champaign); Ha
 rikrishna Tummalapalli (Argonne National Laboratory); Shivam Barwey (Unive
 rsity of Notre Dame); Riccardo Balin and Ramesh Balakrishnan (Argonne Nati
 onal Laboratory); Paul Fischer (University of Illinois Urbana-Champaign, A
 rgonne National Laboratory); and Sheng Di and Franck Cappello (Argonne Nat
 ional Laboratory)\n\nLarge-scale computational fluid dynamics (CFD) simula
 tions routinely generate terabytes of data, making I/O and storage a domin
 ant bottleneck for post-hoc analysis and data-driven workflows. This chall
 enge is amplified on modern GPU-accelerated systems, where the cost of dat
 a movement and checkpointing can rival or exceed computation. We present a
  solver-integrated compression framework for high-order CFD that combines 
 a portable compressed data representation, an embarrassingly parallel I/O 
 pipeline, and a discretization-aware analysis of error propagation. The ap
 proach is implemented in the spectral element solver \texttt{nekRS} using 
 Blosc2 for lossless compression and SZ3 for error-bounded lossy compressio
 n, while remaining applicable to a broader class of high-order discretizat
 ions. A central contribution is a quantity-of-interest (QoI)-aware analysi
 s that accounts for spectral element interpolation, differentiation, and g
 eometric mappings. This framework provides practical guidance for selectin
 g compression tolerances based on workflow requirements, distinguishing be
 tween interpolation-dominated, derivative-sensitive, and projection-based 
 QoIs. We evaluate the approach across representative workflows: (i) jet-in
 -crossflow simulations for visualization, (ii) turbulent channel flow for 
 statistical analysis, (iii) reduced-order modeling (ROM), and (iv) graph n
 eural network (GNN) training. Results show that moderate lossy compression
  achieves substantial data reduction and I/O speedups while preserving key
  QoIs, whereas derivative-based quantities require stricter tolerances con
 sistent with the proposed analysis.These findings demonstrate that compres
 sion can be used in a principled, workflow-aware manner, enabling scalable
  and portable data management for exascale CFD and data-driven scientific 
 computing.\n\nSession Chair: Marta Garcia-Gasulla (Barcelona Supercomputin
 g Center)\n\n
END:VEVENT
END:VCALENDAR