Abstract: However well we protect our systems, there is always a chance they will be penetrated. Constructing practical distributed systems that achieve their goals even after being penetrated is a challenge. During the last few years, there has been considerable progress in the design of intrusion-tolerant (Byzantine) replication systems. The current state of the art can perform well on small scale systems that are usually confined to local area networks (under some assumptions).

This talk presents the first hierarchical Byzantine replication architecture tailored to systems that span multiple wide area sites, each consisting of several replicas. The new architecture dramatically improves system performance (latency and throughput), availability, and manageability, for the price of extra hardware. This work is a first step toward closing the practicality gap between fault-tolerant replication and intrusion-tolerant replication over wide area networks.

Short Bio: Yair Amir is a professor of computer science at Johns Hopkins University, heading the Distributed Systems and Networks lab (www.dsn.jhu.edu). His research goal is to understand the challenges, invent algorithms, and construct software tools that enable high performance, robust, secure and survivable distributed systems. Yair is the creator of the SMesh wireless backbone (www.smesh.org), the Spines overlay network (www.spines.org), the Wackamole and Backhand N-way failover and load balancing projects (www.backhand.org), and the Spread and Secure Spread toolkits (www.spread.org). These systems are deployed in thousands of mission critical systems, are included in commercial products, and are used for research and teaching in universities and research labs around the world.