FASTMath

"The FASTMath SciDAC Institute develops and deploys scalable mathematical algorithms and software tools for reliable simulation of complex physical phenomena and collaborates with application scientists to ensure the usefulness and applicability of FASTMath technologies.  The focus of our work is strongly driven by the requirements of DOE application scientists who work extensively with mesh-based, continuum-level models or particle-based techniques. The FASTMath team has extensive experience developing algorithms and software tools available for such simulations.

Our current research efforts are focused on developing a full range of technologies to improve the reliability, accuracy, and robustness of application simulation codes. In particular, we are developing mesh-based methods including structured, unstructured, and mapped multiblock techniques. We will use embedded boundaries along with high-quality, high-order mesh representations to capture complex geometry, and adaptive mesh refinement to control error in multi resolution simulations. We are developing fast solvers, load-balancing methods, and new remapping techniques. We are focusing effort on robust linear, nonlinear, and eigensolvers.

One of the key challenges facing the scientific computing community is the shift to multi-/manycore nodes and million-way parallelism. Thus a pervasive theme in our work is understanding the most effective ways to implement our algorithms at scale on these architectures, with particular emphasis on hybrid programming models, architecture-aware partitioning and data layout techniques, and communication-reducing algorithms."

https://fastmath-scidac.llnl.gov/index.html

Software Catalog - https://fastmath-scidac.llnl.gov/software-catalog.html

http://www.quest-scidac.org/ 

QUEST Software - http://www.quest-scidac.org/software/

http://super-scidac.org/ 

SUPER Software - http://super-scidac.org/node/392