The objective of this post is to help people to choose the system for a Big Data deployment. I discussed some architectural differences that I think are relevant for making this choice, but didn’t discuss the current performance of the systems much, because I think it’s irrelevant for making this choice.