LINK
Terrific article on MapReduce
After reading this you’ll see just how biased the previous article was. They clearly ignored obvious ways in which to improve the Hadoop benchmarks. Just goes to show that it doesn’t matter how many authors you have contributing to an article, but how smart they are. In fact, I think most of the readers are smarter than Stonebraker, who clearly thinks that his thinly veiled marketing posts are taken seriously.