FerriteCon 2025 Dennis Ogiermann: Order Matters

From FerriteCon 2025, the Ferrite.jl user & developer conference held on August 28, 2025 at the Technical University of Denmark (DTU). In this talk, Dennis Ogiermann (Ruhr-Universität Bochum) explores how element ordering and graph-based sorting techniques can significantly improve the performance of finite element simulations, especially on modern memory-bound computing architectures. Finite element simulations often involve traversing mesh elements in an order determined by general-purpose mesh generators such as Gmsh. These orders can appear nearly random, which leads to poor cache reuse when accessing nodal data repeatedly. By reordering elements so that neighboring elements are visited consecutively — effectively solving a linear arrangement problem on the mesh’s connectivity graph — it becomes possible to improve data locality and reduce memory traffic. In this session, Dennis demonstrates how the Gecko.jl package can be used to compute such optimized orderings and shows benchmark results indicating how graph sorting speeds up simulation runs. The core idea is to minimize a functional that reflects neighborhood relationships in the mesh, thus maximizing reuse of nodal information during assembly and solve loops. This presentation is relevant for scientific computing practitioners, numerical analysts, and developers interested in performance optimization, mesh data structures, and leveraging the Julia ecosystem to build more efficient finite element workflows. Be sure to like, subscribe, and watch other presentations from FerriteCon 2025 for the latest in Julia-based FEM research and tools. #FerriteCon2025 #Ferritejl #JuliaLang #FiniteElementMethod #PerformanceEngineering #GraphAlgorithms #SimulationSpeed Join the channel to support the Julia Language project and access awesome perks! https://www.youtube.com/channel/UC9IuUwwE2xdjQUT_LMLONoA/join For more info on the Julia Programming Language, follow us on Twitter: https://twitter.com/JuliaLanguage and consider sponsoring us on GitHub: https://github.com/sponsors/JuliaLang 00:00 Welcome! 00:10 Help us add time stamps or captions to this video! See the description for details. Want to help add timestamps to our YouTube videos to help with discoverability? Find out more here: https://github.com/JuliaCommunity/YouTubeVideoTimestamps Interested in improving the auto generated captions? Get involved here: https://github.com/JuliaCommunity/YouTubeVideoSubtitles