PCA vs SVD.

Thiru

I m working on classification problem and would like to reduce the features of data.  while using rapidminer,  I see two separate option for SVD and PCA.   I see in many blogs -  PCA and SVD are one and the same, except that SVD is used for non square matrices.  My data is   1000 x 7 dimension.   with combination of few algorithms, SVD and PCA  operators in rapidminer gives big difference in results.

Also , is there any way to have intutive understanding of PCA vs SVD operators used in rapidminer.

thanks.

regards
thiru