Hadoop in Practice: Includes 85 Techniques

very well written

This book is way too old, it's only really good for giving you ideas of what to try. The versions of all the software that this book is based on are often times impossible to obtain now, many of the hadoop sites are in the habit of removing old versions of their projects so you can't download them. For example, the book is primarily based on using hadoop 0.20 and/or cloudera's hadoop version three. It's no longer possible to download those, if you have an old version then hold on to it in case you need it.None of the installation instructions in the book will work with the newer versions of applications. In some cases the entire idea of how you would run and use a tool has changed. Also, the entire way that HDFS and Map-reduce works has changed since YARN was added, so the books explanation of that is old.The book often omits important details like which jar you need to use for a particular piece of code. Classpath and dependency issues are always a nightmare to deal with and the book offers little help with this. He should list everything that you would put in a maven dependency. He often omits the import lines in java code, so you have little idea which class he is referring to in the code.There are often times when he requires you to use software written by him, such as the "File Slurper" that Alex wrote. I am very wary of using any code like that, if it doesn't have the support of the apache/hadoop community then it's very likely to be out of date and unsupported sooner or later. I skipped any chapter I saw like that. I kept seeing this reference to a bash script called "run.sh" in the book, and could not figure out what he was referring to. I could find no such shell script in any software I downloaded. I think it must be a bash script in his git project, like I said I don't want to depend on any code that is not supported by the community.There were also COUNTLESS compatibility issues I found when I tried to do anything. Almost no two pieces of hadoop software work together out of the box. It's so bad that using anything besides cloudera's hadoop was practically impossible. I am not a stupid guy either.Here is my advice to you:1. Use cloudera's pre-built CDH VM, at least at first. I used the CDH 4.5 pre-built VM, and that is the only thing I got to work.2. Do not follow any installation instructions in the HIP book3. Do not follow any installation instructions on the hadoop websites4. Only follow installation/re-configuration instructions found in Cloudera's manual for CDH 4.5 installation5. Do not deviate your configuration from what is norm. For example, I encountered a lot of bugs when I tried switching to java 7.5. You might want to hold off from buying this book until a newer issue is released6. If you use maven for dependencies, make sure you get your hadoop dependencies from the cloudera repository, not maven central7. Instead of reading the book, just go into each of the hadoop project's websites. Skip their installation instructions like I said before, but try to follow any tutorials you see, and try to practice using everything you read.8. After you figure out how to do everything, only then should you try to install stuff from scratch on a new VM. If you try to set up a VM on your own from the start, all the frustration will kill your motivation to learn hadoop.The one thing this book was good for was giving me ideas of what things to try, which is why I give it two stars instead of one.

I'm a beginner and have no experience with any hadoop development before. Before I started to read this book, I had read some articles on the internet. Also read "the hadoop in action". This book really has everything I want to know and have very detailed explanation about the concepts. Every parts in the book are well organized and well written. The author's writing skill is really very amazing, he explains the very complicated concepts in a very plain and easy to understand way. No hypos. The examples in the book are also very practical and very useful and touched many different areas of where hadoop could be used. Not like the online tutorials only play with word count game. The examples in this book are real and practical!

This is a great book. If you want to learn how to actually use Hadoop to do real things, this is the book you need. It also provides great information on HDFS, Bid Data, Data Science, and other topics. The best way to use this book is to get the PDF version and, if you are lucky enough to have two monitors, have the book open on one and a terminal open on the other to your Hadoop environment. That way you can follow along and easily run the code samples. Get this book now if you want to learn Hadoop today!

Probably the first book I have encountered to have specific example of running Hadoop, The beginning is a good lesson about how the book is laid out, then it goes into detail's about getting Hadoop started. I appreciate the author trying to have a go at it, but to update the audience because Hadoop is stilling morphing, there are many caveats about how it is installed via which version of Hadoop, the book had good detail's about Cloudera's version but please do not use this book to have a prescriptive way of installing hadoop as it has changed. The example are good enough to get you started and the detail about what the code means is why the third star. I just think that because Hadoop is still going through metamorphosis stages, all books are just antidotal, many changes have happened during the writing of this book. What is promising it has a version for YARN.

Good

I'm a Hadoop beginner. If you are too, this is not the book for you. The author has stuffed a bit about every conceivable Hadoop-related application of every related Apache project into this book and, beyond being a useful overview discovery guide, it is simultaneously too wide and too shallow (or brief) to be useful to those without substantial knowledge of many of the topics discussed. I'm sure some of the optimization material is useful to those with well-formed questions but this format does not lend itself well to a reader's bootstrap effort on the subject of Hadoop. In all fairness, it does not claim to. Just sayin'.

I read only the first 2 chapters, and got a good fundamental idea of Hadoop File System. Enough to engage in intelligent discussions at work, where I am a software developer.

Only for experienced users with deep knwledge on the Hadoop API. 5 chapters with differents "recipes" to treat out of the ordinary situations, interesting the comments on the not so well known API internals. Buy first Hadoop the definitive guide, and when you master it, get this one