JuRSE Code of the Month - December 2025

Each month we highlight a code from Forschungszentrum Jülich and this month's code is ChASE, co-developed at the Jülich Supercomputing Centre (JSC).

ChASE
The Chebyshev Accelerated Subspace Eigensolver (ChASE) is a modern and scalable library based on subspace iteration with polynomial acceleration to solve dense Hermitian (symmetric) and pseudo-Hermitian algebraic eigenvalue problems. The library is fully parallelised and is particularly effective for sequences of eigenproblems as they often arise in electronic structure theory.

JuRSE selected highlights
Evolving from an internal work horse to a community code, this code made it to an explicit shout-out during the PoF-4 review ("breakthrough", "superior performance", "widely adopted by the scientific community"). The developers do a great job showcasing their achievements and advocating for their code in the right ways and the right places. Besides writing great code, too, of course. Their experience with ChASE led to the ExaNLA initiative, a moonshot project for establishing an open-source, high-performance, scalable, and portable dense numerical linear algebra package ready for Exascale systems.

More information
Website: https://chase-library.github.io/ChASE/
RSD: https://helmholtz.software/software/chase
GitHub: https://github.com/ChASE-library/ChASE