Cactus is a general, modular, parallel environment for solving systems of partial differential equations. The code has been developed over many years by a large international collaboration of numerical relativity and computational science research groups and can be used to provide a portable platform for solving any system of partial differential equations.
Hoard is a scalable memory allocator (malloc replacement) for multithreaded applications. Hoard can dramatically improve your application's performance on multicore machines. No changes to your source are necessary; just link it in. Hoard scales linearly up to at least 64 processors. Supported platforms include Linux, Solaris, Mac OS X, and Windows.
LinAl was designed to bring together C++ and FORTRAN. At the same time LinAl is supposed to be easy to use, fast, and reasonably safe. The LinAl library is based on STL techniques and uses STL containers for the storage of matrix data and STL algorithms where feasible. Low level, algebraic operators, linear solvers, and eigenvalue solvers are implemented, based on calls to BLAS, LAPACK, and CGSOLX.
The Chombo package provides a set of tools for implementing finite difference methods for the solution of partial differential equations on block-structured adaptively refined rectangular grids. Both elliptic and time-dependent modules are included. Support for parallel platforms and standardized self-describing file formats are included.