LAPACK++ is an object-oriented C++ extension to the LAPACK library. Traditionally, linear algebra libraries have been available only in Fortran. However, with an increasing number of programmers using C and C++ for scientific software development, there is a need to have high-quality numerical libraries to support these platforms as well. LAPACK++ provides the speed and efficiency competitive with native Fortran codes, while allowing programmers to capitalize on the software engineering benefits of object-oriented programming.
LAPACK++ supports various matrix classes for vectors, non-symmetric matrices, symmetric positive definite matrices, symmetric matrices, banded, triangular, and tridiagonal matrices; however, the current version does not include all of the capabilities of original Fortran 77 LAPACK. Emphasis is given to routines for solving linear systems consisting of nonsymmetric matrices, symmetric positive definite systems, and solving linear least-square systems. Future versions of LAPACK++ will support eigenvalue problems and singular value decompositions as well as distributed matrix classes for parallel computer architectures. For a more detailed description of the design of LAPACK++, please see [36]. This paper, as well as an Installation manual and Users' Guide are available on netlib. To obtain this software or documentation send a message to netlib@ornl.gov of the form:
send index from c++/lapack++Questions and comments about LAPACK++ can be directed to lapackpp@cs.utk.edu.