AWS re:Invent 2014 | (BAC310) Building Enterprise-Class Backup & Archive Storage Solution Using AWS

In this session, Druva's technical founder shares how the company has used AWS to bring scalable and efficient data backup and archival services to market at scale. Druva has built a scalable storage-as-a-service platform in the cloud, designed to collect, preserve, and help enterprises easily discover information masquerading as ''just some data.'' Druva leveraged Amazon S3 for a large pool of distributed storage for warm copies of data, Amazon Glacier for long-term preservation, and a metadata layer on top of Amazon DynamoDB for high transaction and fast data deduplication. The system is uniquely designed like a file system, so developers can write an application with the same basic assumptions. Druva used the design principles of metadata and data separation and deployed the concept of ''microthreads'' to counter latency challenges. The result is a completely distributed cloud-oriented architecture that is geodistributed, highly available, and able to scale to manage massive volumes of data across multiple customers and geographies. Even if your attention is not particularly on backup solutions, it will be instructive to learn how Druva is taking advantage of these AWS services at massive scale. This session is targeted at developers and at technical and product leaders interested in ways to use AWS in enterprise environments. Sponsored by Druva.