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:20260615T072748Z
LOCATION:Bldg. 6 - Room 002
DTSTART;TZID=Europe/Stockholm:20260701T090000
DTEND;TZID=Europe/Stockholm:20260701T093000
UID:submissions.pasc-conference.org_PASC26_sess149_msaSC109@linklings.com
SUMMARY:Next-Generation Digital Twins: C++-Based Solvers Across CPUs and G
 PUs
DESCRIPTION:Christophe Coreixas (Beijing Normal - Hong Kong Baptist Univer
 sity, University of Geneva) and Jonas Latt (University of Geneva)\n\nDigit
 al Twins are rapidly becoming a key driver of industrial innovation in sec
 tors such as automotive and aeronautics, placing stringent demands on comp
 utational physics solvers for performance, portability, and long-term main
 tainability. Meeting these demands requires more than efficient algorithms
 : it requires strict decoupling of numerical solver logic from parallel ex
 ecution and memory layout.\n\nModern C++ provides practical tools to suppo
 rt this approach. Recent advances, including parallel algorithms and execu
 tion policies, allow developers to express parallelism in a hardware-agnos
 tic way, enabling a single code base to target CPUs and GPUs efficiently. 
 When combined with modular software architectures that isolate numerical m
 odels from hardware-specific concerns, these features deliver performance 
 portability while preserving code clarity, robustness, and flexibility.\n\
 nThe talk illustrates these principles through experiences with high-order
  PDE solvers, multiphysics coupling, and large-scale Digital Twins using a
 daptive mesh refinement and Lattice Boltzmann methods. By highlighting con
 crete design patterns and lessons learned, it focuses on best practices th
 at achieve near-peak performance today while keeping Digital Twin software
  maintainable, modular, and ready for evolving HPC architectures.\n\nDomai
 n: Engineering, Computational Methods and Applied Mathematics\n\nSession C
 hair: Dominik Obrist (University of Bern)\n\n
END:VEVENT
END:VCALENDAR