NAG Library for SMP & Multicore 

NAG Library for SMP & Multicore is a commercial numerical algorithm library developed to harness the huge performance gains from the shared memory parallelism of SMP and multicore processors. Many of the algorithms, or routines, in the Library are specifically tuned to run on multi-socket and multicore systems. Specialized multicore functionality features include dense and sparse linear algebra, FFTs, random number generators, partial differential equations, interpolation, curve and surface fitting, correlation and regression analysis, multivariate methods and sparse iterative eigensolvers (based on ARPACK).