OpenStack Swift: Using, Administering, and Developing

Book Preface

Modern object storage began when applications started to store and serve up assets for web and mobile users. More and more applications needed to deliver content directly to users regardless of whether they were using a laptop, tablet, or smartphone. Object storage is powerful because it can speak the language of the web: HTTP. Additionally, because of its adeptness at storing non-transactional data, it is perfect for archives, backups, file sharing, scientific and research data, and digital media. This book covers both the Swift object storage project (referred to simply as Swift) and SwiftStack, an object storage software company and platform. SwiftStack offers userfriendly ways of accessing, deploying, and managing Swift clusters. Our goal is to show you how to use Swift on your own, present SwiftStack’s main features, and help you decide whether SwiftStack makes sense for you. We believe SwiftStack is a terrific product suitable for many uses, but rest assured that the Swift ecosystem is rich and growing, so there are plenty of options for service providers if SwiftStack turns out not to be a good fit. In the end we want to sell you on Swift object storage, not necessarily on SwiftStack.

Now, six years out from having first used object storage for my applications, I want to share my experiences dealing with the dramatic growth of data. I was fortunate to have a front-row seat as part of the team building one of the first widely used Platform-as-a Service (PaaS) environments running on top of Amazon Web Services. This helped the industry see how powerful an HTTP-based object storage system could be. Personally, I have seen how object storage delivers more flexibility, less lock-in, better control, and lower costs than what traditional storage systems could provide. I have also heard similar things from other organizations. With data (but not IT budgets) growing exponentially, there is absolute certainty that data storage challenges will persist and grow, affecting pretty much everyone who stores and serves data at scale.