https://matthewbonanni.github.io/tags/hpc/Recent content in Hpc on Matthew BonanniHugoen-usMon, 05 Oct 2026 00:00:00 +0000https://matthewbonanni.github.io/posts/2026-10-05-mallard/Mon, 05 Oct 2026 00:00:00 +0000https://matthewbonanni.github.io/posts/2026-10-05-mallard/<div style="text-align: center; margin-bottom: 1.5rem;"> <a href="https://matthewbonanni.github.io/mallard/" style="display: inline-flex; align-items: center; gap: 0.5rem; padding: 0.5rem 1.2rem; border: 1px solid var(--border); border-radius: 6px; text-decoration: none; color: var(--primary); font-size: 0.95rem;">Mallard →</a> </div> <p><a href="https://matthewbonanni.github.io/mallard/">Mallard</a> is an unstructured, GPU-accelerated, high-order finite volume CFD solver I&rsquo;ve been building. It&rsquo;s written in C++20 with Kokkos, so the same code runs on NVIDIA and AMD GPUs as well as multicore CPUs, and it scales across nodes with MPI. Results are bitwise identical on any number of GPUs or ranks.</p>