SBIR Phase II: CohortFS: A Replicated, Parallel Storage System for Cloud Computing

This Small Business Innovation Research (SBIR) Phase II Project advances CohortFS, a replicated, parallel storage platform for cloud computing. CohortFS offers unique capabilities for data partitioning, secure data access and retention, and flexible management of storage at very-large scale, with seamless federation across geographically dispersed, separately-managed data centers. CohortFS improves the flexibility and simplicity of cloud storage management and facilitates uniform management of distributed volumes, whether local, remote, or in the cloud. The CohortFS model for transparent encryption protects data from storage providers and others in shared data center facilities, a key consideration for cloud data privacy and security which are consistently found to be the primary barriers to adoption of Infrastructure-as-a-Service (IaaS) and cloud storage. The broader impact/commercial potential of this project includes advancement of standards-based storage infrastructure, advancement of both open source and proprietary storage technologies, and development of high-value software infrastructure that can be readily commercialized. CohortFS innovations in wide area replication and flexible data placement improve storage performance and manageability in public and private clouds. Using the cloud as an enabling medium, CohortFS translates advances in petascale data organization and secure, wide-area replication to broader markets and commercial sectors. By improving the economy and utility of storage at cloud scale, CohortFS widens the applicability of cloud computing, speeds its adoption, and deepens its impact?goals strongly supportive of the broader NSF mission.