Geeks With Blogs

Arthur Zubarev Compudicted

To the best of my knowledge this is the only book on the market that covers the four major relational database engine’s data storage implementations and namely indexes. I must stress the words data engine because this book deeply covers only one aspect of a relational database – the data management and storage improvements from the indexing prospective, nothing else.

But do not be deceived by its limited content – the indexing part is the most complex, yet a cornerstone part to shipping a successful software product which unfortunately remains a complete black box to most, even seasoned developers! Yet, the book covers in depth.

To expand more, indexing is an often overlooked or even ignored aspect, or at best a post deployment task which typically gets triggered by customers calling in to express how slow the application performance is.

For a software shop then to introduce changes to an application accessing a database poorly at a stage when it was just fully tested and signed off then deployed to production will cost a whole lot more money and headaches than if it were still in early stages of development, not to mention how little time typically is left to make it fast before the customers pull out!

I must tell, I totally agree with Markus – he has correctly stated in the book that it should be the developer, not the Database Administrator who is responsible for the database indexes.

I hope I have enticed you already to get a copy of this book. I personally liked it a lot, especially the second part of it.

I need to also convey to whoever reads my review - the ORM implementations have received its fair portion of the blame, which I am happy Markus had the courage to bring up. Read the book to find out more.

Overall, I admire how the author succeeded to remain neutral in tone and details by not expressing any bias toward any of the four big database vendors.

Even in light of literally dozens of NoSQL (key-value pairs) databases popping up on the landscape today, the book remains very much to the point and actual because the influx of the webscale or sensor data is definitely taking its tall on poorly performing queries.

If somebody would ask me to sum up the book review in a few words I would say this is the book to help you choose the proper RDBMS for your project. If you are already tied to a specific database engine then read this book as soon possible before you develop against it too far to avoid costly performance mistakes.

One suggestion I have to Markus is to remedy the omission on a relatively new feature - the Columnstore Index (a SQL Server 2012 feature). This type of an index promises unforeseen data retrieval speeds and gets even improved further in SQL Server 2014.

Nevertheless, 5 out of 5 from me.

You can order the book directly from

Posted on Monday, November 18, 2013 4:57 PM | Back to top

Comments on this post: SQL Performance Explained by Markus Winand, Book Review

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

Copyright © Compudicted | Powered by: