Parallel Computing for Data Science with Examples in R and Beyond

This is a draft of the first half of a book to be published in 2014 under the Chapman & Hall imprint. Corrections and suggestions are highly encour-aged! c © 2013 by Taylor & Francis Group, LLC. Except as permitted under U.S. copyright law, no part of this book may be reprinted, reproduced, transmit-ted, or utilized in any form by an electronic, mechanical, or other means, now known or hereafter invented, including photocopying, microfilming, and recording, or in any information storage or retrieval system, without written permission from the publishers. 2 Preface Thank you for your interest in this book. I’ve very much enjoyed writing it, and I hope it turns out to become very useful to you. To set the stage, there are a few general points of information I wish to present. Goals: This book hopefully will live up to its title—Parallel Computing for Data Science. Unlike almost every other book I’m aware of on parallel comput-ing, you will not find a single example here dealing with solving partial differential equations and other applications of physics. This book really is devoted to applications in data science—whether you define that term to be statistics, data mining, machine learning, pattern recognition, analytics, or whatever.1 This means more than simply that the book’s examples involve applications chosen from the data science field. It also means that the data structures, algorithms and so on reflect this orientation. This will range from the classic “n observations, p variables ” matrix format to time series to network graph models to various other structures common in data science. While the book is chock full of examples, it aims to emphasize general principles. Accordingly, after presenting an introductory code example in Chapter 1 (general principles are meaningless without real examples to tie them to), I devote Chapter 2 not so much as how to write parallel code, as to explaining what the general factors are that can rob a parallel program of speed. This is a crucial chapter, referred to contantly in the succeeding chapters. Indeed, one can regard the entire book as addressing the plight of the poor guy described at the beginning of Chapter 2: 1Granted, increasingly data science does have some cnnnections to physics, such as in financial modeling and random graphs, but the point is that this book is about data, not physics.