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:20260724T151408Z
LOCATION:Bldg. 8 - B 101
DTSTART;TZID=Europe/Stockholm:20260701T090000
DTEND;TZID=Europe/Stockholm:20260701T110000
UID:submissions.pasc-conference.org_PASC26_sess155@linklings.com
SUMMARY:MS4G - Julia for HPC: Enabling Co-Design in Scientific Workflows
DESCRIPTION:Access the recording\n\nOrganizer(s): Ludovic Räss (University
  of Lausanne, ETH Zurich), Samuel Omlin (ETH Zurich / CSCS), and Michael S
 chlottke-Lakemper (University of Augsburg)\n\nThe fifth instalment of the 
 Julia for HPC PASC minisymposium explores how the Julia language enables c
 o-design to shape scientific workflows that can adapt to rapidly evolving 
 computing architectures. As scientific models grow more complex and comput
 ing moves toward exascale, building trust in HPC software and results beco
 mes essential. Such trust relies on correctness, reproducibility, and reli
 able performance across diverse platforms. This minisymposium highlights h
 ow Julia and its ecosystem support these goals by fostering close co-desig
 n between scientific applications, HPC tools, and emerging hardware and AI
  technologies. Julia’s single-language approach combines ease of use with 
 high performance, allowing scientists and HPC experts to collaboratively d
 evelop, test, and optimise code without separating prototyping from produc
 tion. A central theme is portability across architectures: Julia enables a
  single codebase to target CPUs, GPUs, and novel accelerators, as demonstr
 ated by applications such as Oceananigans.jl and emerging TPU support via 
 Reactant.jl, helping prepare scientific software for future HPC systems. E
 xpert speakers will discuss how Julia enables portable, large-scale Earth 
 system simulations, integrates modern AI compiler technologies, and suppor
 ts differentiable modelling. The minisymposium targets both experienced Ju
 lia users and non-Julia users interested in reproducible and reliable HPC 
 workflows.\n\nOceananigans.jl: GPU-Accelerated Ocean Modeling through Juli
 a-Enabled HPC\n\nAccurate ocean modeling is critical for climate projectio
 ns, yet turbulent motions governing ocean transport remain unresolved in m
 ost climate models, approximated instead through empirical parameterizatio
 ns, a persistent source of bias. Resolving these dynamics globally has bee
 n computationally out...\n\n\nSimone Silvestri (Politecnico di Torino, Mas
 sachusetts Institute of Technology); Gregory Wagner (Aeolus Labs, Massachu
 setts Institute of Technology); Navid Constantinou (University of Melbourn
 e); Valentin Churavy (University of Augsburg); and Raffaele Ferrari (Massa
 chusetts Institute of Technology)\n---------------------\nReactant.jl: Opt
 imize Julia Functions for High Performance on CPU, GPU, TPU\n\nScientific 
 models are today limited by compute resources, forcing approximations driv
 en by feasibility rather than theory. They consequently miss important phy
 sical processes and decision-relevant regional details. Advances in AI-dri
 ven supercomputing — specialized tensor accelerators, AI comp...\n\n\nWill
 iam Moses (University of Illinois Urbana-Champaign; Google)\n-------------
 --------\nHierarchical Precision and Recursion for Accelerating Symmetric 
 Linear Solves on MXUs\n\nSymmetric positive-definite system solvers based 
 on Cholesky factorization are a critical performance bottleneck in large s
 cientific workflows, such as climate modeling. Addressing this demands co-
 design across algorithms, software abstractions, and emerging hardware, pa
 rticularly as modern AI accel...\n\n\nVicky Carrica (Massachusetts Institu
 te of Technology)\n---------------------\nMulti-Physics Geophysical Flow S
 imulations Using JustRelax.jl\n\nSimulating the solid Earth's thermo-mecha
 nical evolution requires solving coupled, non-linear Stokes and heat-diffu
 sion problems across large domains with sharp material contrasts, complex 
 non-linear visco-elasto-plastic rheologies, and long time scales. Traditio
 nally, these simulations run on High-...\n\n\nPascal S. Aellig (Johannes G
 utenberg University Mainz), Albert de Montserrat (ETH Zurich), Ludovic Räs
 s (University of Lausanne), Boris J. P. Kaus (Johannes Gutenberg Universit
 y Mainz), and Christian Schuler (University of Lausanne)\n\nDomain: Climat
 e, Weather, and Earth Sciences, Physics, Computational Methods and Applied
  Mathematics\n\nSession Chairs: Ludovic Raess (University of Lausanne, ETH
  Zurich) and Samuel Omlin (ETH Zurich / CSCS)
END:VEVENT
END:VCALENDAR