Some systems are closed and proprietary while others are open and free. A fine example of the latter is OpenStack, a set of free, open-source software tools for building and managing very scalable public and private clouds. OpenStack enables organizations of all sizes to avoid the risky lock-in inherent with proprietary cloud solutions. It’s been called both “the Linux of the cloud” and “the future of cloud computing.”
OpenStack was created in 2010 in part by NASA and is now backed by a large, engaged community of developers and many leading companies. Its modular architecture includes components for compute, networking, and storage.
Of particular interest is OpenStack’s support for both object and block storage. Known as Swift, OpenStack Object Storage is a redundant storage system that uses clusters of standardized servers and can scale to petabytes of data. Rather than a traditional file system, it provides distributed storage for static data like “virtual machine images, photo storage, email storage, backups and archives.” The solution stores objects and files to multiple drives, thus delivering redundancy and data protection. If a server or drive fails, OpenStack Object Storage will replicate the data to other nodes in the cluster. Thanks to this functionality, users can deploy commodity drives and servers rather than costlier, high-end arrays. Additionally, scaling horizontally means just adding new servers to the cluster.
Many organizations, of course, use block storage and they have the option of OpenStack Block Storage, also known as Cinder. The system offers persistent block-level storage devices for use with OpenStack compute instances. Cinder is useful when data must be rapidly accessed, such as for database applications, because it can access specific locations on a drive. It can use local Linux servers for storage or storage platforms from a variety of vendors. Cinder also manages the creation, attaching, and detaching of block devices to servers, as well as snapshot management for backing up data.