Jimmy Lin

After you've read a basic MapReduce book like the O'Reilly Hadoop guide, this should be your next step. Even if you're not doing text processing per se, this the best (and so far only) published book I've seen about advanced MapReduce programming. Lin and Dyer start from the supposition that MapReduce is its own programming paradigm on a par with, say, OOP or functional programming and then show how to apply this mindset to various big data problems, starting with various iterations of word show more count and moving on to sophisticated algorithms like page rank, graph search and expectation maximization. This book has clear and in-depth descriptions of the partitioning and sorting techniques you can employ to use MapReduce to its fullest potential. show less