![ETHOS logo](/sites/default/files/styles/front_page_card/public/ethos-llnl-project-card.png?itok=xJ1CQwbN)
ETHOS
The Enabling Technologies for High-Order Simulations (ETHOS) project performs research of fundamental mathematical technologies for next-generation high-order simulations algorithms.
![MFEM logo on a blue background](/sites/default/files/styles/front_page_card/public/mfem-llnl-project-card.png?itok=3Q3CImbi)
MFEM
The open-source MFEM library enables application scientists to quickly prototype parallel physics application codes based on PDEs discretized with high-order finite elements.
![gollnlp logo of interconnected circles and lightning bolt](/sites/default/files/styles/front_page_card/public/gollnlp-llnl-project-card.png?itok=h_4r9Pau)
GOLLNLP
Responding to a DOE grid optimization challenge, an LLNL-led team developed the mathematical, computational, and software components needed to solve problems of the real-world power grid.
![Tarik Dzanic](/sites/default/files/styles/front_page_card/public/2023-09/tarik-dzanic-432x285.png?itok=Y0mZbZ_I)
Tarik Dzanic
As Computing’s seventh Fernbach Fellow, postdoctoral researcher Tarik Dzanic will develop new algorithms and test them in computational physics simulations under the mentorship of Bob Anderson.
![Andrew Gillette](/sites/default/files/styles/front_page_card/public/2023-08/Andrew-Gillette-432x285.png?itok=9n9oCaIU)
Andrew Gillette
CASC computational mathematician Andrew Gillette has always been drawn to mathematics and says it’s about more than just crunching numbers.
![Brian Gunney playing guitar at a campsite](/sites/default/files/styles/front_page_card/public/2023-06/brian-gunney-guitar-leaderboard.png?itok=H0vZqxrU)
Brian Gunney
From wind tunnels and cardiovascular electrodes to the futuristic world of exascale computing, Brian Gunney has been finding solutions for unsolvable problems.
![concentric circles where the top half is yellow and orange and the bottom half is blue, set on a background of dark green at bottom and light green at top](/sites/default/files/styles/front_page_card/public/2024-06/marbl-solvers-comp-news.png?itok=PDQYmYJU)
Matrix unloaded: GPU-boosted solvers for diffusion physics
Developed by LLNL and Portland State University researchers, innovative matrix-free solvers offer performance gains for complex multiphysics simulations.
![A flow chart showing the classification of the types of errors in scientific computing.](/sites/default/files/styles/front_page_card/public/2024-06/errors_classification_flowchart_432x285.png?itok=TXf8Z7A5)
Symposium paper formally verifies whether linear systems will converge
A new method defines a formal specification for convergence, which can be used to derive a set of machine-checkable conditions to guarantee a convergent solution to a differential equation.