Viewstamped Replication: A New Primary Copy Method to Support Highly-Available Distributed Systems
PODC '88 Proceedings of the seventh annual ACM Symposium on Principles of distributed computing
Lazy replication: exploiting the semantics of distributed services
PODC '90 Proceedings of the ninth annual ACM symposium on Principles of distributed computing
Lazy replication: exploiting the semantics of distributed services (extended abstract)
ACM SIGOPS Operating Systems Review
References to remote mobile objects in Thor
ACM Letters on Programming Languages and Systems (LOPLAS)
Lazy replication: exploiting the semantics of distributed services
EW 4 Proceedings of the 4th workshop on ACM SIGOPS European workshop
Hi-index | 0.00 |
One possible advantage a distributed system has over a centralized system is the ability to move objects from one node to another. For example, we may want to move an object if the node where it resides is overloaded. This thesis proposes to use a location service to aid in finding objects that move. The service is highly-available; it will tolerate system failures like node crashes and network partitions without shutting down completely. The service is also efficient; the response time of the service is reasonable, and it does not increase the number and sizes of messages excessively. We achieve high availability and efficiency by replicating the service state. The replication technique we have chosen is a new method, the multipart timestamp technique that is based on multipart timestamps and gossip messages. This technique provides us with higher availability and efficiency than traditional replication techniques without sacrificing consistency. We also extend this technique to allow reconfiguration.