[BioC] FYI from apple employee RE: [R] Memory error in Mac OS X

[David Williamson]

Stefano,

I saw this thread and would like to know how the tiger/64bit 
R/bioconductor compile is coming along.  I'm interested in using the 
hopach package to run memory intensive jobs that are currently limited 
by the OSX 10.8 memory ceiling.

Have you been able to use  the new TIGER kernel or the new Xtools 2.0 
libraries to compile R in 64-bit, or are you waiting for the release?

Are there another threads or newsgroups dedicated to this topic?

Thanks for your assistance.

~david




Feb 24, 2005, at 1:43 AM, Betty Gilbert wrote:

 > Please ignore this is you're not interested in what my friend who
 > works for apple had to say regarding my issue which is explained in
 > detail at the bottom of this exchange...
 >
 > me:
 >  The data from this simulation has to be processed at several stages
 > of data processing so at some point I do have to have to make a
 > dissimilarity matrix from this large matrix, as well as ones that I
 > have filtered.
 >
 > appleguy:
 >>>> Your matrix is too big; adding more memory, or killing off other
 >>>> processes won't help.  R is clearly not very smart when it comes to
 >>>> dealing with large chunks of data, and you're running into a
 >>>> fundamental upper limit on the amount of data that it can deal
 >>>> with.
 >>>>
 >>>> The only hope you have of dealing with this is to find a 64-bit
 >>>> system (many Unix boxes fit this description) and use the text or
 >>>> X-windows based R there.  If you have a login on a Unix machine at
 >>>> work that has R installed, try your program there.  In a few
 >>>> months, when Tiger is out and about, the G5 Macs will also be able
 >>>> to run 64-bit R.
 >

and we are going to release a G5 Tiger compliant version of R indeed
after Tiger will be officially released. We are working on this.
stefano

 >
 > me:
 >>> Yeah I was unsuccesful trying to download the unix version of r last
 >>> night(which is also the windows version). Christopher said it needed
 >>> the fortran compiler(?). According to takao the linux we use doesn't
 >>> have any gui capability so I'm not sure I can make the dendograms
 >>> I'm gonna need if I installed R on it. It Linux 2.6.7. and is some
 >>> version of red hat(?). Oh well...
 >
 > appleguy:
 >
 >> Being Linnex, it is probably running on a PC, so it won't be a 64-bit
 >> system anyway.   What sort of timeframe do you have on this? I might
 >> be able to track down a 64-bit system, but it may take a little
 >> while.
 >>>
 > If time is tight, then you should ask around to see a) whether anyone
 > has a 64-bit unix machine; i.e. a Sun, SGI or Alpha box, and b)
 > whether they know how to generate the graphs without the GUI (I
 > suspect the GUI is not required to generate them, only view them).
 > Until then, you're going to need to reduce the size of your dataset
 > I'm afraid. 8(
 >
 >
 > OG message to rhelp:
 >
 > I'm sorry if the answer to my problem is buried in the archives. I
 > have limited experience with R  and I couldn't find a solution to my
 > particular problem. I am running  Mac OS X Aqua GUI v1.01 on a new G5
 > running os 10.3.8 with a 1.8Ghz processor and 1GB of sdram. I just
 > downloaded bioconducter a week ago and I'm trying to cluster a matrix
 > I created with a simulation with dimensions
 > dim(nca35)
 > [1] 10481    12
 >
 > with size
 >>  object.size(nca352)
 > [1] 1426204
 >
 > I checked my ulimits variable on the unix terminal and it says it's
 > unlimited as does
 >>  mem.limits()
 > nsize vsize
 >    NA    NA
 > But I'm still getting errors like the following with funtions in the
 > cluster package
 >>  daisy(nca352, metric= "euclidean", stand=FALSE)->dnca35
 > Error: cannot allocate vector of size 858213 Kb
 > *** malloc: vm_allocate(size=878813184) failed (error code=3)
 > *** malloc[599]: error: Can't allocate region
 > if it helps i also checked
 >>  gc()
 >          used (Mb) gc trigger   (Mb)
 > Ncells 448662 12.0     741108   19.8
 > Vcells 847630  6.5  135357901 1032.7
 >
 > I tried the suggested unix command in the memory help doc but that
 > doesn't work in the Aqua GUI. Can someone tell me how to change the
 > Vcells? Although to the best of my understanding (which is limited) I
 > shouldn't have to do that. Any suggestions would be greatly
 > appreciated.
 > thanks,
 > betty
 >
 > --
 > Betty Gilbert
 > lgilbert at berkeley.edu
 > Taylor Lab
 > Plant and Microbial Biology
 > 321 Koshland Hall
 > U.C. Berkeley
 > Berkeley, Ca 94720
 >
 > _______________________________________________
 > Bioconductor mailing list
 > Bioconductor at stat.math.ethz.ch
 > https://stat.ethz.ch/mailman/listinfo/bioconductor





~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
David Williamson
The J. David Gladstone Institutes
Gladstone Institute of Cardiovascular Disease
1650 Owens Street, room 571 Y2
San Francisco, CA 94158

UCSF mailbox 1230

Work:   	415.734.4945 (voice mail 6136)
Fax:    	415.285.5632
Home:	http://dww.ucsf.edu/~david/
Email: 	dwilliamson at gladstone.ucsf.edu
	[[alternative text/enriched version deleted]]