Skip to main content

Survivable Distributed Storage: From Theory to Reality

Researchers: Greg Ganger, Manish Prasad

Research Area: Survivable Distributed Systems

Abstract

Scope: Military decision makers need distributed storage that can provide timely access to accurate information even in the face of attacks and failures. This project tackles technical concerns keeping COTS providers from producing such systems.

Outcomes: 1. A compelling case for COTS providers to develop survivable storage solutions. 2. Protocols for efficient, versatile, scalable survivable distributed systems. 3. Protocols and techniques that can be applied to other survivable distributed systems to increase scalability and efficiency. 4. A prototype survivable storage service deployed at Carnegie Mellon and used to store real data for CyLab researchers. 5. Release the prototype system developed as free software for others to use in research, class projects or perhaps their own deployment.