[Gate-users] Voxelized phantom source

Gottwald, Jennifer jennifer.gottwald at rmp.uhn.ca
Tue Dec 20 14:11:27 CET 2016


Hi David,



Thank you for your great E-Mail.



I am using my own code ( see attachment and the error screenshot). In general the code is running with and without the job splitter but so slowly on CPU that I need to speed it up. It took 3.5 weeks for 846MB run splitted in 5 jobs. Ram 7.8GB, on local disk, CPU 2000.000 MHz.



Therefore I investigated the options to set cuts but this doesn't work. I get the message that it can't find my phantom when I enable the cuts. I am also not sure how I can find out what the range in mm for each of these cuts is (Geant user guide says 100keV for 1 mm - is that correct?).



Best wishes and thank you for all your help!



Jennifer



-----Original Message-----

From: Gate-users [mailto:gate-users-bounces at lists.opengatecollaboration.org] On Behalf Of David Boersma

Sent: Saturday, December 17, 2016 8:56 AM

To: gate-users at lists.opengatecollaboration.org<mailto:gate-users at lists.opengatecollaboration.org>

Subject: Re: [Gate-users] GATE: Voxelized source/phantom speed issue on CPU and error cut off values



Hi Jennifer,



I am confused. In the first sentence you mention that you have speed issues, which implies that it is running normally, only slow. But in the last sentence you mention that you get an error message (missing

phantom) which implies it is *not* running normally, the run gets aborted already during the initialization phase. I must be misunderstanding your question.



I have not yet used the job splitter myself (the one in "cluster_tools"). But I would recommend to first try running things first without the job splitter, and to verify that the output is correct. Only after you are confident that you are getting correct results you should try to make it faster. (And after you did that, you need to check again that the result is still correct...)



One way to get errors regarding the missing phantom is by forgetting to download the examples and benchmark data. In order to get them you need to set the corresponding flags from OFF to ON during the cmake step.

(This is assuming that you are indeed running one of the examples. If you run your own code on your own data then you do not need these example/benchmark data.)



Some practical questions:



(1) Which script are you trying to run? If it is not one of the examples, could you send your "mac" file to this mailing list?



(2) Are you sure that the input data is present on your system?

Specifically, if you are running e.g.

examples/example_PHANTOM_SOURCE/Voxelized_Phantom_Source/VoxelizedPhantom.mac,

which uses brain_phantom.h33, could you check that brain_phantom.h33 exists and is not a broken link?



(3) If you still get errors after check (2): please quote the full error output.



(4) If you do not get errors any more, Gate produces a sensible output file but it's just slow: please tell us how long it took, and on what kind of a system you are running (how much RAM, what kind of CPU with what kind of clock speed, do you run everything on a local disk or are there network disks involved?).



Good luck,

David Boersma



Den 15/12/2016 kl. 16:36, skrev Gottwald, Jennifer:

> Hi,

>

>

>

> I am a new user of GATE 7.2 and have speed issues on running PET

> voxelized source/phantom simulations on CPU

>

>

>

> I already use the option to split jobs.

>

>

>

> Does someone has experience with that? I would be very grateful for

> any advice

>

>

>

> I try to set cut of values in the physics mac but I get error messages

> that the voxelized phantom can't be found.

>

>

>

> Thanks

>

> Jennifer

>

>

>

>

> This e-mail may contain confidential and/or privileged information for

> the sole use of the intended recipient.

> Any review or distribution by anyone other than the person for whom it

> was originally intended is strictly prohibited.

> If you have received this e-mail in error, please contact the sender

> and delete all copies.

> Opinions, conclusions or other information contained in this e-mail

> may not be that of the organization.

>

>

>

> _______________________________________________

> Gate-users mailing list

> Gate-users at lists.opengatecollaboration.org<mailto:Gate-users at lists.opengatecollaboration.org>

> http://lists.opengatecollaboration.org/mailman/listinfo/gate-users

>

_______________________________________________

Gate-users mailing list

Gate-users at lists.opengatecollaboration.org<mailto:Gate-users at lists.opengatecollaboration.org>
http://lists.opengatecollaboration.org/mailman/listinfo/gate-users

This e-mail may contain confidential and/or privileged information for the sole use of the intended recipient. 
Any review or distribution by anyone other than the person for whom it was originally intended is strictly prohibited. 
If you have received this e-mail in error, please contact the sender and delete all copies. 
Opinions, conclusions or other information contained in this e-mail may not be that of the organization.
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.opengatecollaboration.org/mailman/private/gate-users/attachments/20161220/55d88a78/attachment-0001.html>
-------------- next part --------------
A non-text attachment was scrubbed...
Name: ERROR.png
Type: image/png
Size: 14115 bytes
Desc: ERROR.png
URL: <http://lists.opengatecollaboration.org/mailman/private/gate-users/attachments/20161220/55d88a78/attachment-0001.png>
-------------- next part --------------
A non-text attachment was scrubbed...
Name: master.mac
Type: application/octet-stream
Size: 2147 bytes
Desc: master.mac
URL: <http://lists.opengatecollaboration.org/mailman/private/gate-users/attachments/20161220/55d88a78/attachment-0002.obj>
-------------- next part --------------
A non-text attachment was scrubbed...
Name: physics.mac
Type: application/octet-stream
Size: 1475 bytes
Desc: physics.mac
URL: <http://lists.opengatecollaboration.org/mailman/private/gate-users/attachments/20161220/55d88a78/attachment-0003.obj>


More information about the Gate-users mailing list