SPOTRF crashing for matrices greater than 40Kx40K

Open discussion for MAGMA

SPOTRF crashing for matrices greater than 40Kx40K

Postby raffayhamid » Sat Jan 11, 2014 5:07 pm

UPDATE 2: By following the instructions given the the link I mentioned in 'update 1', it seems that the problem has been resolved (just ran spotrf for an 80k x 80k matrix).

UPDATE 1: After some more search, I found this link that seems to shed some light on the problem:

Any extra information would still be appreciated!

Hi --

My understanding is that the magma_spotrf() function has an "out-of-core" implementation, which I assume means that the size of my problems should be upper-bounded by the host memory, and not my device memory (please correct me if needed).

When I run the (provided by MAGMA) testing_spotrf.cpp file however, the code breaks at magma_malloc_cpu() for matrices with sizes > 40Kx40K (I have reason to believe that the same behavior is true for magma_malloc_pinned() function). I'm curious what is the reason for this apparent barrier.

One possible explanation that might make sense is that perhaps to exploit the modern instruction-set, the MAGMA CPU memory is 32-bit aligned ( see point # 9), which might be the reason for this problem I'm encountering ? Any explanation would be much appreciated.

Is there a way around this issue?


PS: It's interesting to see that the graph shown on slide 12 of this presentation ( ... MA_1.4.pdf) ends around 46K. After running some more experiments on my end I found that I can run the test till 46K as well, and not more. Seems like there must be a more principled reason for this limit.

PPS: I'm getting more convinced that this has something to do with the 32bit virtual address space. Just for kicks, note that 2^31(/46000/46000) > 1 while 2^31(/47000/47000) < 1, which seems peculiar in this context (even though it uses only 31 bits and not all 32 -- maybe 1 bit is being used for something else...?).
Posts: 3
Joined: Sun Jan 05, 2014 2:32 am

Re: SPOTRF crashing for matrices greater than 40Kx40K

Postby mgates3 » Tue Jan 14, 2014 11:54 am

Yes, for large matrices, 64-bit sizes are needed. The include demonstrates enabling this with MKL.
Posts: 442
Joined: Fri Jan 06, 2012 2:13 pm

Return to User discussion

Who is online

Users browsing this forum: Google [Bot] and 2 guests