Our paper "On the robustness of high-order upwind summation-by-parts methods for nonlinear conservation laws" has been published in the Journal of Computational Physics.

We tested the numerical code Trixi.jl on the Jülich Supercomputing Centre and ran it at 61,440 MPI ranks with excellent scaling.

We have submitted our new paper "Secure numerical simulations using fully homomorphic encryption".

We would like to invite you to the upcoming?CAAPS Workshop?on?Particle-based numerical simulations?which will take place at the Centre for Advanced Analytics and Predictive Sciences of the University of Augsburg on 5th and 6th November 2024.

You can find further information here. Looking forward to seeing you there! ??

Valentin Churavy has joined the High-Performance Scientific Computing Lab on August 1st, 2024 as a postdoctoral researcher.

Using the numerical simulation framework Trixi.jl we implement adaptive multiphysics coupling mechanisms. With that we can simulate domains that make use of different kinds of physics. They are coupled through the boundaries in a way that, in principle, one could couple very different systems. Thanks to our adaptive model selection, we can prescribe criteria for changing model adaptively.

The HPSC Lab has created two packages for using fully homomorphic encryption with the Julia programming language.

Two members of the HPSC Lab gave talks at this year's JuliaCon conference taking place in Eindhoven, Netherlands from July 9th through 13th.

Simon Candelaresi has joined the High-Performance Scientific Computing Lab on July 1st, 2024 as a postdoctoral researcher.

2D single-phase simulation of the liquefaction of the Julia logo with TrixiParticles.jl.

Our paper "Multirate time-integration based on dynamic ODE partitioning through adaptively refined meshes for compressible fluid dynamics" has been published in the Journal of Computational Physics.

Ioana Lupu has joined the High-Performance Scientific Computing Lab on June 1st, 2024 as the new team assistant.