FFPACK: finite field linear algebra package

The FFLAS project has established that exact matrix multiplication over finite fields can be performed at the speed of the highly optimized numerical BLAS routines. Since many algorithms have been reduced to use matrix multiplication in order to be able to prove an optimal theoretical complexity, this paper shows that those optimal complexity algorithms, such as LSP factorization, rank determinant and inverse computation can also be the most efficient. Categories and Subject Descriptors G.4 [Mathematical Software]: Algorithm design and analysis; F.2.1 [Analysis of Algorithms and Problem Complexity]: Numerical Algorithms and Problems—computations in finite fields. General Terms Algorithms, Experimentation, Performance. Keywords Word size Finite fields; BLAS level 1-2-3; Linear Algebra Package; Matrix Multiplication; LSP Factorization

• Type: Conference
• Year: 2004
• Where: ISSAC
• Authors: Jean-Guillaume Dumas, Pascal Giorgi, Clément Pernet
• Keywords: ISSAC 2004, Lsp Factorization, Matrix Multiplication, Optimal Complexity Algorithms