MAP-JOIN-REDUCE: Toward Scalable and Efficient Data Analysis on Large Clusters | IEEE Journals & Magazine | IEEE Xplore