Michael Kane

Agreed. The `sub.big.matrix` was implemented with row and column offsets in the `BigMatrix` class. We could probably do a lot more by implementing a new accessor class. However, we need...

bigalgebra assumes BLAS/LAPACK and both take contiguous column-major c-arrays as input. If your `sub.big.matrix` is simply a subset of the columns then there is no problem since the memory is...

Thanks very much! Please note that deepcopy does let you specify columns and rows to copy from a big.matrix object.

@renozao do you create named, shared mutexes when nmf is performed in parallel? If so, and R crashes it is possible that they could not be cleaned up correctly. @rronen...

Can you run the following in a terminal window so we can see the shared resources? ``` ls -lah /dev/shm ``` If you see resources that end with _counter then...

Shoot, I can verify this. This is my mistake. Now, let's see if we can get the example to work, with the memory leak. Please exit R, remove all of...

OK, I think I found the problem and it's mine... Give me an hour or so and I'll report back.

@rronen you probably need to remove all of the associated shared resources. The `big.matrix` objects are not being cleaned up but I'm having trouble figuring out why this might be...

OK, I think I found the problem. The example uses "doMC" which forks for parallel processes. When I run ``` r library(bigmemory) library(doMC) registerDoMC(cores=2) a = big.matrix(nrow=3, ncol=3, shared=TRUE) desc...

So, just to summarize it looks like there are 3 separate issues. 1. nmf hangs - @renozao and I have not been able to reproduce this using parallel and/or shared...