[Insight-developers] Why is VariableLengthVector::GetSquaredNorm so slow?

[Bradley Lowekamp]

Hello David,

It looks like you have factored the GetSize() method out side of the loop. This likely removes branching and exceptions from the loop making it suitable for SIMD/SEE vectorization.

Brad

On Sep 14, 2011, at 9:01 AM, David Doria wrote:

> I have noticed that if I loop over the elements of two
> VariableLengthVectors to compute the squared norm of their difference,
> it is about 3x faster than if I use the GetSquaredNorm() function. I
> have attached a demo which uses itkTimeProbe to show this. Can anyone
> explain the difference? My ManualDifference2 seems to me to be
> identical to GetSquaredNorm.
> 
> Thanks,
> 
> David
> <GetSquaredNormSpeedTest.cxx>_______________________________________________
> Powered by www.kitware.com
> 
> Visit other Kitware open-source projects at
> http://www.kitware.com/opensource/opensource.html
> 
> Kitware offers ITK Training Courses, for more information visit:
> http://kitware.com/products/protraining.html
> 
> Please keep messages on-topic and check the ITK FAQ at:
> http://www.itk.org/Wiki/ITK_FAQ
> 
> Follow this link to subscribe/unsubscribe:
> http://www.itk.org/mailman/listinfo/insight-developers

========================================================
Bradley Lowekamp  
Lockheed Martin Contractor for
Office of High Performance Computing and Communications
National Library of Medicine 
blowekamp at mail.nih.gov


-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://www.itk.org/mailman/private/insight-developers/attachments/20110914/4ff9c5a8/attachment.htm>