Bug#910002: ITP: combblas -- an extensible parallel graph library for graph analytics
Package: wnpp
Severity: wishlist
Owner: Drew Parsons <dparsons@debian.org>
* Package name : combblas
Version : 1.6.2
Upstream Author : (c) 2018, The Regents of the University of California, through Lawrence Berkeley National Laboratory and University of California, Santa Barbara
* URL : https://people.eecs.berkeley.edu/~aydin/CombBLAS/html/index.html
* License : BSD
Programming Lang: C++, C
Description : an extensible parallel graph library for graph analytics
The Combinatorial BLAS (CombBLAS) is an extensible distributed-memory
parallel graph library offering a small but powerful set of linear
algebra primitives specifically targeting graph analytics.
- The Combinatorial BLAS development influences the Graph BLAS
standardization process.
- It achieves scalability via its two dimensional distribution and
coarse-grained parallelism.
- CombBLAS powers HipMCL, a highly-scalable parallel implementation of
the Markov Cluster Algorithm (MCL).
- Operations among sparse matrices and vectors use arbitrary user
defined semirings.
combblas implements AWPM, a row permutation algorithm now offered in
superlu-dist as an alternative to the non-free MC64 algorithm.
Consequently combblas will make superlu-dist much more useful than it
has been in the past.
It will be packaged under the Debian Science team alongside
superlu-dist.
Reply to: