Geeks With Blogs

Arthur Zubarev Compudicted

Unlocking Scientific Data

It is probably yesterday’s news that Python is the de facto programming language for anything Data Science. And the latest book on Python and HDF5 integration is a more recent proof to that.

I want to state here that it seems to be the ONLY book on the market today on the becoming increasingly popular self contained data storage and manipulation format HDF5 that explains how to program against it in Python at an enterprise level.

Even though it is a book review, let me briefly explain that HDF5 is a database like, hierarchical file structure closely resembling the early file-based databases implementing Balanced Tree indexing for fast data retrieval. The fact the file is self contained helps keep data, attributes and even computational results together for transparent data exchange, in fact it is so inter-operating platforms exchange-ready it takes complete care of the platform differences as little-endian versa big-endian for example, and boy Andrew knows how to explain that in the book!

Actually, the book has made me aware of how important it is to use proper technologies when you have no idea where (here platform) your data will be consumed.

As a brief side note, myself I programmed hierarchical data structures for fats data retrieval in the early 90s, in C, not even knowing they are called B-Trees. And the concept has such a broad implementation.

So in short, the book is excellent, written in a concise, professional manner (between me and you, 0 volume inflating fluff).

The author has made sure the book is full of useful examples covering each nuance or an important feature so reading this book feels natural and logical. I am also glad the author devoted a significant effort to convey to a developer ( I hate the word ‘programmer’ :-) ) on the proper methods of concurrent programming, which is what a pity – a common omission in many beginners’ books.

I am sure this book will make you going or will let you start coding against HDF5 in no time. I am sure this book will be used as a table reference (or on your computer desktop).

I am giving this book a 5 out of 5 rating, kudos to O’Reilly that has delivered yet another outstanding publication.

Disclaimer: This book was given to me for free as part of the blogger review program by O'Reilly Media.

Posted on Wednesday, November 27, 2013 7:14 PM | Back to top

Comments on this post: Python and HDF5 - Unlocking Scientific Data By Andrew Collette, O'Reilly Media, Inc. Book Review

No comments posted yet.
Your comment:
 (will show your gravatar)

Copyright © Compudicted | Powered by: