On Tue, Jul 17, 2012 at 12:08 PM, Douglas Bates <ba...@stat.wisc.edu> wrote: > On Tue, Jul 17, 2012 at 10:44 AM, French, Joshua > <joshua.fre...@ucdenver.edu> wrote: >> Thank you all for the responses. >> >> Christian, I didn't know about the copy_aux_mem option. I will have to >> take a look at that. >> >> Dirk, thanks for looking into the 64-bit matrix indices. >> >> Doug, the place in my code where I get the error is when I multiply >> matrices. I might have matrices X and Y, where X is 300000x500 and Y is >> 500x300000 and I want Z = X * Y. > > You will need to come up with another algorithm. The size of Z is > 3,000,000 by 3,000,000 and storing this as a dense matrix will require > about 67 terabytes of memory > >> (3000000 * 3000000 * 8) / 2^30 > [1] 67055.23
To be fair, it's 300k by 300k (not 3mill by 3mill), so: R> (300000 * 300000 * 8) / 2^30 [1] 670.5523 So, I guess 0.67 terabytes ... no problemo ;-) -steve -- Steve Lianoglou Graduate Student: Computational Systems Biology | Memorial Sloan-Kettering Cancer Center | Weill Medical College of Cornell University Contact Info: http://cbio.mskcc.org/~lianos/contact _______________________________________________ Rcpp-devel mailing list Rcpp-devel@lists.r-forge.r-project.org https://lists.r-forge.r-project.org/cgi-bin/mailman/listinfo/rcpp-devel
