MBLAS: Modular Basic Linear Algebra Subprograms, Design and Speedup Techniques
نویسندگان
چکیده
Computation in prime fields Zp is used by various algorithms in computer algebra, and plays an important role in practice, especially when p is represented direct by machine datatype. The application area is wide spreaded, but in most cases, major part of the calculation can be reduced to a combination of basic linear operations, as in the calculations with polynomials over Zp. Arithmetic in Zp is very simple and the reduction by mod p costs significantly. To achieve practical superior efficiency, detailed analysis and careful tuning is required, especially when treating vectors and matrices, a bulk of homogeneous data. With this recognition and the necessity for sharable and reusable efficient subprograms in mind, we started a few years ago a project to develop a program library of modular version of BLAS, basic linear algebra subprograms. This poster gives an overview of our design, and describes some practical and successful techniques used in our implementation, as well as showing experimental results. There is a similar software by Dumas et al. [1]. Our effort will be characterized by the following features.
منابع مشابه
An Extended Set of Fortran Basic Linear Algebra Subprograms
This paper describes an extension to the set of Basic Linear Algebra Subprograms. The extensions are targeted at matrix-vector operations which should provide for efficient and portable implementations of algorithms for high performance computers. An Extended Set of Fortran Basic Linear Algebra Subprograms Jack J. Dongarra † Mathematics and Computer Science Division Argonne National Laboratory ...
متن کاملAccelerating resolution-of-the-identity second-order Møller-Plesset quantum chemistry calculations with graphical processing units.
The modification of a general purpose code for quantum mechanical calculations of molecular properties (Q-Chem) to use a graphical processing unit (GPU) is reported. A 4.3x speedup of the resolution-of-the-identity second-order Møller-Plesset perturbation theory (RI-MP2) execution time is observed in single point energy calculations of linear alkanes. The code modification is accomplished using...
متن کاملDesign of Linear Algebra Libraries for High Performance Computers
This paper discusses the design of linear algebra libraries for high performance computers. Particular emphasis is placed on the development of scalable algorithms for MIMD distributed memory concurrent computers. A brief description of the EISPACK, LINPACK, and LAPACK libraries is given, followed by an outline of ScaLAPACK, which is a distributed memory version of LAPACK currently under develo...
متن کاملTechnical Paper Accepted for Publication in Siam Review Software Libraries for Linear Algebra Computations on High Performance Computers 1 Software Libraries for Linear Algebra Computations on High Performance Computers
This paper discusses the design of linear algebra libraries for high performance computers. Particular emphasis is placed on the development of scalable algorithms for MIMD distributed memory concurrent computers. A brief description of the EISPACK, LINPACK, and LAPACK libraries is given, followed by an outline of ScaLAPACK, which is a distributed memory version of LAPACK currently under develo...
متن کامل