Parallel rank calculation for sparse matrices -- suggestions?
8 visualizaciones (últimos 30 días)
Mostrar comentarios más antiguos
Barry
el 30 de Ag. de 2022
Comentada: Bruno Luong
el 6 de Sept. de 2022
I need to calculate the rank of large (> 1 terabyte of non-zero elements) sparse matrices with MATLAB. Exploring the Parallel Toolbox, but can't seem to find anything that convinces me what is offers will be helpful. If I'm wrong, can someone point me in the right direction? Ideally, I'd just want to take my existing code that has "r=sprank(A)" in it and have that library call run in parallel, perhaps with some additional annotation as needed. Would using a GPUARRAY help here, for example? Doesn't seem like it, but perhaps I'm wrong. Hoping someone here can help me out. Thanks!
Barry Fagin
Professor of Computer Science
barry.fagin@afacademy.af.edu
9 comentarios
Matt J
el 5 de Sept. de 2022
A = sparse(tall,skinny)
How tall? How skinny? What currently rules out the approach already discussed below of using rank(full(A.'*A))?
Bruno Luong
el 6 de Sept. de 2022
"After doing a little digging, I now believe sprank() merely counts the number of non-zero columns. This is not what I require."
It does something more intelligent than that it match the row to column through Dulmange Mendelsohn permutation. In your case might be all the non-zero column can be matched. But that is exactly what structural rank means.
Respuesta aceptada
Bruno Luong
el 30 de Ag. de 2022
Movida: Bruno Luong
el 30 de Ag. de 2022
5 comentarios
Bruno Luong
el 31 de Ag. de 2022
IMO the doc should clarify this.
But then I still have some doubt. If user runs multiple sprank in parallel, it will take less worker to work on one sprank. So at the end do we really speed up the whole thing?
Edric Ellis
el 1 de Sept. de 2022
Yes, the "extended capability" description could probably do with some refinement to make it clear exactly what works. (Today, there are a good number of MATLAB functions that cannot be run on a thread pool worker, and this "extended capability" really means simply that the function can run on a thread pool worker).
The performance situation is ... complicated, unfortunately. There is no single simple answer as to whether running N copies of a given function concurrently on workers is faster than running N copies sequentially on the client. It depends on all sorts of details of the implementation of that function - but primarily whether the function is already intrinsically multithreaded by MATLAB itself.
Más respuestas (2)
Matt J
el 30 de Ag. de 2022
Is the matrix square or is it tall and thin? If the later, then it may be an easier computation to compute rank(A.'*A).
Bruno Luong
el 5 de Sept. de 2022
For thin and tall sparse matrix A of size (m x n), m>>n and n in the order of 1000s, it might be possible to compute the rank using q-less qr, which is better than rank(A'*A) which has the drawback of square the condition number.
A=sprand(100000,10,0.1)*sprand(10,100,0.1);
R=qr(A,0);
rank(full(R))
0 comentarios
Ver también
Categorías
Más información sobre Sparse Matrices en Help Center y File Exchange.
Community Treasure Hunt
Find the treasures in MATLAB Central and discover how the community can help you!
Start Hunting!