[Rtk-users] Issue with cudaConstantVolumeSource in RTK for Large Volume Sizes

[Simon Rit]

Plese use the mailing list
<https://www.creatis.insa-lyon.fr/mailman/listinfo/rtk-users> (cc). Can you
provide the error message when you execute this code? You'd need to catch
and display the exception, e.g., with TRY_AND_EXIT_ON_ITK_EXCEPTION
<https://github.com/RTKConsortium/RTK/blob/master/include/rtkMacro.h#L106>:
TRY_AND_EXIT_ON_ITK_EXCEPTION(reconstructionFilter->Update())

There are of course limitations linked to the CPU and GPU memory. Your
volume is about 8 GB, I don't know what is the size of your projection
images. There are a few solutions to workaround this:
- only read a few projections at a time by setting the subset size
<https://github.com/RTKConsortium/RTK/blob/master/applications/rtkfdk/rtkfdk.cxx#L148>,
For it to be efficient, don't read the full stack beforehand as done here
<https://github.com/RTKConsortium/RTK/blob/master/applications/rtkfdk/rtkfdk.cxx#L65>
.
- use the StreamingImageFilter to work on parts of the volume at once
<https://github.com/RTKConsortium/RTK/blob/master/applications/rtkfdk/rtkfdk.cxx#L202>
.

Simon

On Fri, May 16, 2025 at 8:26 AM 胡继涛 <hujitao226 at 163.com> wrote:

>
> I hope this message finds you well.
>
> I am currently using RTK for reconstruction on a system equipped with an
> NVIDIA RTX 4090 GPU (24 GB VRAM). When attempting to use the
> cudaConstantVolumeSource to create a large volume (e.g., 1000×1000×1000), I
> encounter an error during reconstruction。
>
> Could you kindly advise on what might be causing this issue? Are there any
> known limitations or considerations regarding volume size in this context?
>
> Thank you very much for your time and assistance.
>
> Best regards,
> 胡继涛
> hujitao226 at 163.com
>
> <https://dashi.163.com/projects/signature-manager/detail/index.html?ftlId=1&name=%E8%83%A1%E7%BB%A7%E6%B6%9B&uid=hujitao226%40163.com&iconUrl=https%3A%2F%2Fmail-online.nosdn.127.net%2Fqiyelogo%2FdefaultAvatar.png&items=%5B%22hujitao226%40163.com%22%5D>
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://www.creatis.insa-lyon.fr/pipermail/rtk-users/attachments/20250516/a2bca9cb/attachment.htm>