We're sorry but this page doesn't work properly without JavaScript enabled. Please enable it to continue.
Feedback

LIBRSB: Universal Sparse BLAS Library

Formal Metadata

Title
LIBRSB: Universal Sparse BLAS Library
Subtitle
A highly interoperable Library for Sparse Basic Linear Algebra Subroutines and more for Multicore CPUs
Title of Series
Number of Parts
542
Author
Contributors
License
CC Attribution 2.0 Belgium:
You are free to use, adapt and copy, distribute and transmit the work or content in adapted or unchanged form for any legal purpose as long as the work is attributed to the author in the manner specified by the author or licensor.
Identifiers
Publisher
Release Date
Language

Content Metadata

Subject Area
Genre
Abstract
Many science and engineering models reduce to problems involving huge sparse matrices -- matrices where most of the values are zeroes. Such computations are resource-intensive (time, memory, energy), and much research was devoted into data structures ("formats") and algorithms leading to fast sparse matrix operations. Yet, most such formats are highly specialized and seldom make it into a solid software package apt for general use. The RSB (Recursive Sparse Blocks) data structure is a format that addresses performance concerns for current shared-memory multicore CPUs, while also avoiding dead ends in terms of usability. The LIBRSB library implements RSB with all the necessary operations to manipulate sparse matrices in most computations, in the most popular programming languages, and on many hardware platforms. This talk will give an overview of the concepts behind LIBRSB and its main usage modes. Intended audience: Developers of Linear Systems Solvers based on Iterative Methods, or General Computing Packages. Expected prior knowledge: Familiarity in any of C, C++, Fortran, Python, GNU Octave.