Diff for "TestsNewConfigJuly2009" - MRC CBU Imaging Wiki
location: Diff for "TestsNewConfigJuly2009"
Differences between revisions 1 and 8 (spanning 7 versions)
Revision 1 as of 2009-08-04 08:24:33
Size: 39
Editor: RhodriCusack
Comment:
Revision 8 as of 2009-08-04 08:33:34
Size: 3202
Editor: RhodriCusack
Comment:
Deletions are marked like this. Additions are marked like this.
Line 1: Line 1:
Describe TestsNewConfigJuly2009 here. = Configuration =
Rob installed Matlab R2006b (7.3) R2008b and R2009a I think this should
be enough version wise hopefully.

Machines details are as follows

L43 / L44 = CentOS 5.3 x86-64
L45/ L46 = CentOS 5.3 i386

All systems have a 80GB /scratch partition to testing using local disc
during the analyses.

== EEG Lab ==
Tested by Jason Taylor - works fine on 32 & 64 bit machines

== FSL ==
Tested by Martin Monti

CentOs 5 32bit (FSL4.1.4)
2221.62user 51.34system 45:29.18elapsed 83%CPU (0avgtext+0avgdata 0maxresident)k
0inputs+0outputs (41major+16526197minor)pagefaults 0swaps

CentOs 5 64 Bit (FSL4.1.4)
1815.98user 37.20system 32:30.90elapsed 94%CPU (0avgtext+0avgdata 0maxresident)k
0inputs+0outputs (5major+14040943minor)pagefaults 0swaps

Additional testing by Jonathan Peelle

FSL test suite.
32-bit FSL on l46(i386): 37m16.659s
32-bit FSL on l44(x86-64): 37m23.956s
64-bit FSL on l44(x86-64): 29m38.987s

== MNE/Freesurfer ==
Tested by Dan Wakeman

They seem to be missing some key libraries (or possibly improperly installed) e.g. lapack for lots of software. Will play with 3d stuff once this is sorted.

== SPM MEG ==
Tested by Jason Taylor/ Daniel Mitchell
.mex64 required recompile (got spurious "out of memory" error in filearray access) but then appears to work fine

== SPM FMRI ==
aa version 3 test script, all stages up to and including smoothing
Runs fine on 32 & 64 bit machines
Benchmarks




== Software maintainers copied from AvailableSoftware ==

||||||||||<style="text-align: center;">'''Software''' ||
||Package ||<style="vertical-align: top;">First port of Call (i.e. wiki page) ||<style="vertical-align: top;">Manager ||<style="vertical-align: top;">Maintainer ||<style="vertical-align: top;">Contributors ||
||EEG Lab ||<style="vertical-align: top;"> ||<style="vertical-align: top;">Jason Taylor ||<style="vertical-align: top;">Jason Taylor ||<style="vertical-align: top;">Jason Taylor ||
||<style="vertical-align: top;">Fieldtrip ||<style="vertical-align: top;">FieldtripInformation ||<style="vertical-align: top;">Jonathan Peelle ||<style="vertical-align: top;">Jonathan Peelle ||<style="vertical-align: top;"> ||
||Freesurfer ||<style="vertical-align: top;"> ||<style="vertical-align: top;">Daniel Wakeman ||<style="vertical-align: top;">Daniel Wakeman ||<style="vertical-align: top;">Daniel Wakeman ||
||FSL ||<style="vertical-align: top;"> ||<style="vertical-align: top;">Martin Monti ||<style="vertical-align: top;">Martin Monti ||<style="vertical-align: top;"> ||
||<style="vertical-align: top;">MNE ||<style="vertical-align: top;">http://imaging.mrc-cbu.cam.ac.uk/meg/MneAnalysis ||<style="vertical-align: top;">Daniel Wakeman ||<style="vertical-align: top;">Daniel Wakeman ||<style="vertical-align: top;">Daniel Wakeman ||
||<style="vertical-align: top;">SPM || <style="vertical-align: top;">http://imaging.mrc-cbu.cam.ac.uk/imaging/SpmInformation (FMRI) http://imaging.mrc-cbu.cam.ac.uk/meg/SpmAnalysis (MEG/EEG) ||<style="vertical-align: top;">Rhodri Cusack ||<style="vertical-align: top;">Rhodri Cusack ||<style="vertical-align: top;">Daniel Mitchell, Jason Taylor, Rik Henson, Rhodri Cusack ||

Configuration

Rob installed Matlab R2006b (7.3) R2008b and R2009a I think this should be enough version wise hopefully.

Machines details are as follows

L43 / L44 = CentOS 5.3 x86-64 L45/ L46 = CentOS 5.3 i386

All systems have a 80GB /scratch partition to testing using local disc during the analyses.

EEG Lab

Tested by Jason Taylor - works fine on 32 & 64 bit machines

FSL

Tested by Martin Monti

CentOs 5 32bit (FSL4.1.4) 2221.62user 51.34system 45:29.18elapsed 83%CPU (0avgtext+0avgdata 0maxresident)k 0inputs+0outputs (41major+16526197minor)pagefaults 0swaps

CentOs 5 64 Bit (FSL4.1.4) 1815.98user 37.20system 32:30.90elapsed 94%CPU (0avgtext+0avgdata 0maxresident)k 0inputs+0outputs (5major+14040943minor)pagefaults 0swaps

Additional testing by Jonathan Peelle

FSL test suite. 32-bit FSL on l46(i386): 37m16.659s 32-bit FSL on l44(x86-64): 37m23.956s 64-bit FSL on l44(x86-64): 29m38.987s

MNE/Freesurfer

Tested by Dan Wakeman

They seem to be missing some key libraries (or possibly improperly installed) e.g. lapack for lots of software. Will play with 3d stuff once this is sorted.

SPM MEG

Tested by Jason Taylor/ Daniel Mitchell .mex64 required recompile (got spurious "out of memory" error in filearray access) but then appears to work fine

SPM FMRI

aa version 3 test script, all stages up to and including smoothing Runs fine on 32 & 64 bit machines Benchmarks

Software maintainers copied from AvailableSoftware

Software

Package

First port of Call (i.e. wiki page)

Manager

Maintainer

Contributors

EEG Lab

Jason Taylor

Jason Taylor

Jason Taylor

Fieldtrip

FieldtripInformation

Jonathan Peelle

Jonathan Peelle

Freesurfer

Daniel Wakeman

Daniel Wakeman

Daniel Wakeman

FSL

Martin Monti

Martin Monti

MNE

http://imaging.mrc-cbu.cam.ac.uk/meg/MneAnalysis

Daniel Wakeman

Daniel Wakeman

Daniel Wakeman

SPM

<style="vertical-align: top;">http://imaging.mrc-cbu.cam.ac.uk/imaging/SpmInformation (FMRI) http://imaging.mrc-cbu.cam.ac.uk/meg/SpmAnalysis (MEG/EEG)

Rhodri Cusack

Rhodri Cusack

Daniel Mitchell, Jason Taylor, Rik Henson, Rhodri Cusack

CbuImaging: TestsNewConfigJuly2009 (last edited 2013-03-07 21:22:56 by localhost)