Object Storage’s uptake within the industry is increasingly rapidly. OpenStack Swift is the OpenStack variant of Object Storage, and allows you to consume storage by using API calls. Web based storage is becoming more popular in the world of cloud and business’s are leveraging web based storage to transform the way they store and consume data. The most popular use of Object Storage today, would be Amazon’s S3 with other service providers following the trend to redefine how storage is consumed within the cloud.
With the growing interest in OpenStack, and my personal interest growing as well, I recently accepted an opportunity to review a book about OpenStack Swift (Implementing Cloud Storage – Openstack Swift).
This book has been published by Packt Publishing and provides an architecture overview, walk through instructions on how to install, manage and tune OpenStack swift implementations.
Chapter 1 – Cloud Storage – Why Can’t I be like Google?
Chapter 2 – OpenStack Swift Architecture
Chapter 3 – Installing OpenStack Swift
Chapter 4 – Using Swift
Chapter 5 – Managing Swift
Chapter 6 – Choosing the right Hardware
Chapter 7 – Tuning your Swift Installation
Chapter 8 – Additional Resources
Without much prior experience in OpenStack, I decided to follow with the book and get under the covers. Chapters 1 & 2 cover a good background of cloud storage and openstack architecture, this also covers the additional OpenStack products. I think this section of the book could have more detail around the use cases for cloud storage and diagrams of how the architecture fits together in more detail (high level diagrams provided only).
Flowing onto chapter 3, the installation of OpenStack Swift, this is a ‘follow the bouncing ball’ procedure. I think this section covered off the installation procedures well, could have done with some more detail around the commands being used and how the installation can be done at scale.
Chapter 4 covers the usage of Swift, this goes into detail about different clients and how to interact with them. To my surprise the book covers a wider variety of ways to interact and consume Swift storage including Amazon S3 compatibility options than I expected. I think this chapter did a great job at covering the content the author was trying to get across.
The day to day management of Swift is covered in chapter 5, this chapter covers a few products that can be used and also different ways to handle hardware related errors. I think this chapter was very light on content and should have contained more information. It does not provide enough insight into how to monitor the product on a day to day basis and should focus more on operational procedures.
Chapters 6,7 & 8 are self explanatory from the chapter titles, the content in these chapters was very good. Chapter 7 certainly has some really good tips for tuning swift once it is installed. Although I was left wondering where you would use some of the recommendations and whether they are suitable with and without scale, I guess time will tell.
Overall, I thought the book was a great insight into OpenStack Swift, and certainly helped me break the ice on building a Swift environment to play with. I did have to refer to OpenStack Swift official documentation a few times after the reviewing the book to clarify a few items that the book didn’t explain in enough detail for my curiosity. The main key item that I would like to see content about in the book is the networking aspects of how swift operates. No content is provided around recommendations to load balance and or protect the cluster if using it publicly.
I would recommend this books to people who have an understanding of linux, openstack and are looking at object storage. It is a great starting point, but does not cover everything you will need to know to setup Swift in a production environment.