Azzera filtri
Azzera filtri

Parallel rank calculation for sparse matrices -- suggestions?

4 visualizzazioni (ultimi 30 giorni)
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 Commenti
Matt J
Matt J il 5 Set 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
Bruno Luong il 6 Set 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.

Accedi per commentare.

Risposta accettata

Bruno Luong
Bruno Luong il 30 Ago 2022
Spostato: Bruno Luong il 30 Ago 2022
Something I don't get : sprank doc tells when you run witth threadpool it runs in parallel.
  5 Commenti
Bruno Luong
Bruno Luong il 31 Ago 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
Edric Ellis il 1 Set 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.

Accedi per commentare.

Più risposte (2)

Matt J
Matt J il 30 Ago 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).
  9 Commenti
Barry
Barry il 5 Set 2022
Modificato: Barry il 5 Set 2022
I like the A.'*A trick. Or one could ...wait for it ...
... use MATLAB's rref() command, which works on sparse matrices, and then count the nonzero rows. Sigh.
But that in turn begs the question of roundoff errors. Time to move to a new question, which I'll post shortly.
--BF
Matt J
Matt J il 5 Set 2022
Modificato: Matt J il 5 Set 2022
I would take caution before relying on rref. It has been often found to lack robustness, see e.g.,

Accedi per commentare.


Bruno Luong
Bruno Luong il 5 Set 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))
ans = 10

Categorie

Scopri di più su Sparse Matrices in Help Center e File Exchange

Community Treasure Hunt

Find the treasures in MATLAB Central and discover how the community can help you!

Start Hunting!

Translated by