A security architecture for fault-tolerant systems

Process groups are a common abstraction for fault-tolerant computing in distributed systems. We present a security architecture that extends the process group into a security abstraction. Integral parts of this architecture are services that securely and fault tolerantly support cryptographic key distribution. Using replication only when necessary, and introducing novel replication techniques when it was necessary, we have constructed these services both to be easily defensible against attack and to permit key distribution despite the transient unavailability of a substantial number of servers. We detail the design and implementation of these services and the secure process group abstraction they support. We also give preliminary performance figures for some common group operations.