A Deployment Guide for Elastic Storage Object

Due to the explosion of unstructured data generated by individuals and organizations, a new storage paradigm called "Object Storage" has been developed. Object Storage stores data in a flat namespace that scales to trillions of objects. The design of object storage also simplifies how users access data, supporting new types of applications and allowing users to access data by a variety of methods, including mobile devices and web applications. Data distribution and management are also simplified, allowing greater collaboration across the globe.

OpenStack Swift is an emerging open-source object storage software platform that is widely used for cloud storage. Elastic Storage, based on IBM® General Parallel File System (GPFS™) technology, is a high-performance and proven product that is used to store the data for thousands of mission-critical commercial installations worldwide. (Throughout this IBM Redpaper™ publication, Elastic Storage is used to refer to GPFS 4.1.)

Elastic Storage also automates common storage management tasks, such as Tiering and Archiving at scale. Together, Elastic Storage and OpenStack Swift provide an enterprise-class object storage solution that efficiently stores, distributes, and retains critical data.

This paper provides instructions about how to set up and configure Swift with Elastic Storage. It also provides an initial set of preferred practices to ensure optimal performance and reliability.

The goal of this paper is to describe the benefits of using GPFS as the underlying file system with OpenStack Swift, guide the administrator through the installation and configuration of Elastic Storage Object, and describe the general set of configurations and scenarios that have been validated. It is intended for administrators who are familiar with GPFS and OpenStack Swift components.