Dear Doug,
I saw your reply to this question a while ago. I'm using a mac pro with 16G memory but having the same problem. Do you know if this is a matlab memory issue (matlab R2009a) or an overall memory issue of the machine? I'd appreciate if you can give me some helps. (I have 12 runs, each run has 200 TRs and 20 events, TR=2s)
Thanks very much! Brown
ps. Here are the commands and error messages:
mkanalysis-sess -analysis $ANALYSIS -TR 2 -paradigm $PARAFILE -event-related -funcstem $FUNCSTEM -nconditions 9 -mcextreg -runlistfile $RLF -fsd bold -fir 10 30 -native
selxavg3-sess -s $SUBJECTID -df sessdir -analysis $ANALYSIS -no-con-ok
Whitening Computing whitening matrices MATLAB(19078,0xb0987000) malloc: *** mmap(size=1382400000) failed (error code=12) *** error: can't allocate region *** set a breakpoint in malloc_error_break to debug ??? Out of memory. Type HELP MEMORY for your options.
Error in ==> fast_selxavg3 at 603 Sinv = zeros(ntptot,ntptot,flac0.acfbins);
pure virtual method calledOpening log file: /var/root/java.log.19078
Abort ------------------------------------------ ERROR: fast_selxavg3() failed\n
I think you might need a bigger computer. A run of that data set will require almost 1G, and there are places in the code where several data structures of the same size will need to exist simultaneously, so you can easily use 2 or 3 or 4 G to process a single run. In theory you could divide them up into smaller bits as long as you handle the stimulus schedule files properly.
doug
Schadwinkel, Stefan wrote:
*Hi,*>**>*as I tried to show in the posting, I tried it on 3 Machines:*>**>*1) Linux, 1 GB memory: out of memory during OLS Residual Pass, then Java crashes*>*2) Linux, 2 GB memory: out of memory during Whitening*>*3) Mac (Mac OS X 10.5), 4 GB memory: as well: out of memory during Whitening*>**>*Thanks a lot & best regards,*>**>*Stefan Schadwinkel
*
It is "only" trying to allocate 1.3GB, so it might be that matlab does not have enough memory when it runs. Is there a way to give matlab more memory? If not, you can re-make your analysis with -acfbins 10. The default is 30, so this will reduce the needed memory by a factor of 3. I don't think it will affect your results much. doug
Po-Jang (Brown) Hsieh wrote:
Dear Doug,
I saw your reply to this question a while ago. I'm using a mac pro with 16G memory but having the same problem. Do you know if this is a matlab memory issue (matlab R2009a) or an overall memory issue of the machine? I'd appreciate if you can give me some helps. (I have 12 runs, each run has 200 TRs and 20 events, TR=2s)
Thanks very much! Brown
ps. Here are the commands and error messages:
mkanalysis-sess -analysis $ANALYSIS -TR 2 -paradigm $PARAFILE -event-related -funcstem $FUNCSTEM -nconditions 9 -mcextreg -runlistfile $RLF -fsd bold -fir 10 30 -native
selxavg3-sess -s $SUBJECTID -df sessdir -analysis $ANALYSIS -no-con-ok
Whitening Computing whitening matrices MATLAB(19078,0xb0987000) malloc: *** mmap(size=1382400000) failed (error code=12) *** error: can't allocate region *** set a breakpoint in malloc_error_break to debug ??? Out of memory. Type HELP MEMORY for your options.
Error in ==> fast_selxavg3 at 603 Sinv = zeros(ntptot,ntptot,flac0.acfbins);
pure virtual method calledOpening log file: /var/root/java.log.19078
Abort
ERROR: fast_selxavg3() failed\n
I think you might need a bigger computer. A run of that data set will require almost 1G, and there are places in the code where several data structures of the same size will need to exist simultaneously, so you can easily use 2 or 3 or 4 G to process a single run. In theory you could divide them up into smaller bits as long as you handle the stimulus schedule files properly.
doug
Schadwinkel, Stefan wrote:
/Hi,
/>/ />/as I tried to show in the posting, I tried it on 3 Machines: />/ />/1) Linux, 1 GB memory: out of memory during OLS Residual Pass, then Java crashes />/2) Linux, 2 GB memory: out of memory during Whitening />/3) Mac (Mac OS X 10.5), 4 GB memory: as well: out of memory during Whitening />/ />/Thanks a lot & best regards, />/ />/Stefan Schadwinkel /
-- Po-Jang (Brown) Hsieh Neuroscience and Behavioral Disorders Program Duke-NUS Graduate Medical School 8 College Road, Singapore 169857 -- Lab website: Brain & Consciousness Lab https://sites.google.com/site/brainconsciousnesslab/
Freesurfer mailing list Freesurfer@nmr.mgh.harvard.edu https://mail.nmr.mgh.harvard.edu/mailman/listinfo/freesurfer
Hi,
I had the same problem with the matlab 32-bit version. Try to install the (recent) matlab 64-bit on your Mac.
cheers,
Erhan
2011/11/8 Douglas N Greve greve@nmr.mgh.harvard.edu:
It is "only" trying to allocate 1.3GB, so it might be that matlab does not have enough memory when it runs. Is there a way to give matlab more memory? If not, you can re-make your analysis with -acfbins 10. The default is 30, so this will reduce the needed memory by a factor of 3. I don't think it will affect your results much. doug
Po-Jang (Brown) Hsieh wrote:
Dear Doug,
I saw your reply to this question a while ago. I'm using a mac pro with 16G memory but having the same problem. Do you know if this is a matlab memory issue (matlab R2009a) or an overall memory issue of the machine? I'd appreciate if you can give me some helps. (I have 12 runs, each run has 200 TRs and 20 events, TR=2s)
Thanks very much! Brown
ps. Here are the commands and error messages:
mkanalysis-sess -analysis $ANALYSIS -TR 2 -paradigm $PARAFILE -event-related -funcstem $FUNCSTEM -nconditions 9 -mcextreg -runlistfile $RLF -fsd bold -fir 10 30 -native
selxavg3-sess -s $SUBJECTID -df sessdir -analysis $ANALYSIS -no-con-ok
Whitening Computing whitening matrices MATLAB(19078,0xb0987000) malloc: *** mmap(size=1382400000) failed (error code=12) *** error: can't allocate region *** set a breakpoint in malloc_error_break to debug ??? Out of memory. Type HELP MEMORY for your options.
Error in ==> fast_selxavg3 at 603 Sinv = zeros(ntptot,ntptot,flac0.acfbins);
pure virtual method calledOpening log file: /var/root/java.log.19078
Abort
ERROR: fast_selxavg3() failed\n
I think you might need a bigger computer. A run of that data set will require almost 1G, and there are places in the code where several data structures of the same size will need to exist simultaneously, so you can easily use 2 or 3 or 4 G to process a single run. In theory you could divide them up into smaller bits as long as you handle the stimulus schedule files properly.
doug
Schadwinkel, Stefan wrote:
/Hi,
/>/ />/as I tried to show in the posting, I tried it on 3 Machines: />/ />/1) Linux, 1 GB memory: out of memory during OLS Residual Pass, then Java crashes />/2) Linux, 2 GB memory: out of memory during Whitening />/3) Mac (Mac OS X 10.5), 4 GB memory: as well: out of memory during Whitening />/ />/Thanks a lot & best regards, />/ />/Stefan Schadwinkel /
-- Po-Jang (Brown) Hsieh Neuroscience and Behavioral Disorders Program Duke-NUS Graduate Medical School 8 College Road, Singapore 169857 -- Lab website: Brain & Consciousness Lab https://sites.google.com/site/brainconsciousnesslab/
Freesurfer mailing list Freesurfer@nmr.mgh.harvard.edu https://mail.nmr.mgh.harvard.edu/mailman/listinfo/freesurfer
-- Douglas N. Greve, Ph.D. MGH-NMR Center greve@nmr.mgh.harvard.edu Phone Number: 617-724-2358 Fax: 617-726-7422
Bugs: surfer.nmr.mgh.harvard.edu/fswiki/BugReporting FileDrop: www.nmr.mgh.harvard.edu/facility/filedrop/index.html
Freesurfer mailing list Freesurfer@nmr.mgh.harvard.edu https://mail.nmr.mgh.harvard.edu/mailman/listinfo/freesurfer
The information in this e-mail is intended only for the person to whom it is addressed. If you believe this e-mail was sent to you in error and the e-mail contains patient information, please contact the Partners Compliance HelpLine at http://www.partners.org/complianceline . If the e-mail was sent to you in error but does not contain patient information, please contact the sender and properly dispose of the e-mail.
freesurfer@nmr.mgh.harvard.edu
-
Douglas N Greve -
Erhan Genc -
Po-Jang (Brown) Hsieh