Sparse matrix creation in parallel

Question

Are there any algorithms that allow efficient creation (element filling) of sparse (e.g. CSR or coordinate) matrix in parallel?

score 0 · Answer 1 · answered Aug 01 '10 at 12:40

0

If you store your matrix as a coordinate map, any language which has a concurrent dictionary implementation available should do the job for you.

Java's got the ConcurrentHashMap, and .NET 4 has ConcurrentDictionary, both of which allow multi-threaded non-blocking (afaik) element insertion in parallel.

answered Aug 01 '10 at 12:40

tzaman

46,925
11
90
115

Concurrency and parallelism are not the same thing. The problem here is truly data-parallel filling of elements into a sparse matrix. Specifically in my case I want to implement it on GPUs. – Fic Firic Aug 01 '10 at 19:42

score 0 · Accepted Answer · answered Aug 03 '10 at 21:43

There are no efficient algorithms for creating sparse matrices in data-parallel way. Plausible is coordinate matrix type which requires sorting after content filling, but that type is slow for matrix products etc.

Solution is you don't build sparse matrix - you don't keep it in memory; you do implicit operations in place when you're calculating elements of sparse matrix.

Sparse matrix creation in parallel

2 Answers2