Thanks Mark. How did you initiate an additional thread in itk?
Mark Wyszomierski
<markww at gmail.com> To:
"mrcheung at mdanderson.org" <mrcheung at mdanderson.org>
cc:
02/02/2005 12:20 PM
Please respond to Mark
Wyszomierski
Subject:
Re: [Insight-users] re: 3D deformable registration and dual processor
computer
Hello,
I was faced with the same problem in 2D deformable registration. All I
did to harness the extra processor(s) was to spawn a new registration
thread per extra processor(s). This has cut the total processing time
for a dataset as you would expect by the number of extra processors.
I haven't used the 3D deformable registration, maybe you can setup a
similar method if it lends itself to such a setup. I would also be
interested in hearing if there is an 'ITK' way to take advantage of
the extra processors.
Mark
On Wed, 2 Feb 2005 09:45:30 -0600, mrcheung at mdanderson.org
<mrcheung at mdanderson.org> wrote:
> I am thinking of ways to speed up the itk 3D deformable registration with
> hardware solutions. The registration time for CT to CT fusion is in
hours.
> My naive question is: can the use of two processor computer speed up the
> itk 3D deformable registration by a factor of approximately two? Is there
a
> way to use itk deformable registraiton module explicitly to utilize
> multiple processor processing that I may use to speed up our deformable
> registration time?
>
> Thanks a lot, Rex
>
> _______________________________________________
> Insight-users mailing list
> Insight-users at itk.org
> http://www.itk.org/mailman/listinfo/insight-users
>