Abstract
This paper presents a specification for membership services designed to operate in an asynchronous environment that is subject to both crash and omission failures. In case of a loss of connectivity in the system (partition), the specification supports partitionable operation, in which components that (temporarily) disconnected from each other operate autonomously. Furthermore, the framework suggested is achievable in an asynchronous environment, yet offers useful semantics of split-up and merging that provide the application developer with sufficient information to recover a consistent state. By taking into account hidden changes, the framework supports consistent mending of partitions and allows for a full recovery.
Original language | English |
---|---|
Pages | 343 |
Number of pages | 1 |
State | Published - 1996 |
Event | Proceedings of the 1996 15th Annual ACM Symposium on Principles of Distributed Computing - Philadelphia, PA, USA Duration: 23 May 1996 → 26 May 1996 |
Conference
Conference | Proceedings of the 1996 15th Annual ACM Symposium on Principles of Distributed Computing |
---|---|
City | Philadelphia, PA, USA |
Period | 23/05/96 → 26/05/96 |