,
High Performance Recursive Linear Algebra Library

Author: Ivan Šimeček


Keywords
cache utilization, high performance, linear algebra, recursive programming style

Abstract
For good performance of every computer program, the good cache utilization is crucial. In numerical algebra libraries (such as BLAS or LAPACK) is good cache utilization achieved by explicit loop restructuring (mainly loop blocking), but it requires difficult cache behavior analysis. In this paper, we represent recursive implementation of some routines from numerical algebra library. This implementation leads to good cache utilization with no need to analyze cache behavior due to the "natural" partition of data.

What's new
October 1 2009 -  version 1.00 of the High Performance Recursive Linear Algebra Library is released.


History
version 1.00 - the first fully functional version of the High Performance Recursive Linear Algebra Library.


Known bugs


(Possible) future works

Used in papers
Memory Hierarchy Behavior Study during the Execution of Recursive Linear Algebra Library
Recursive implementation of high performance numerical algebra library
Recursive Implementation of High Performance Numerical Algebra Library



Download:
High Performance Recursive Linear Algebra Library 1.00
Example of using the High Performance Recursive Linear Algebra Library Gemm performace



BACK