| Speaker: | William H. Sanders Center for Reliable and High-Performance Computing Coordinated Science Laboratory University of Illinois | |
|---|---|---|
| Topic: | Dependability Management of Distributed Objects in AQuA using Proteus | |
| Date: | Tuesday, February 2, 1999 | |
| Time: | 9:30 AM | |
| Place: | Gould-Simpson, Room 701 |
In this talk, I will provide an overview of the AQuA architecture, which provides an infrastructure for building dependable distributed systems. In particular, the AQuA architecture provides dependable distributed objects through replication, where the type and degree of replication is managed dynamically, depending on the needs of applications. To do this, the AQuA architecture uses the QuO runtime to process availability requests and provide high-level recovery, the Proteus dependability manager to configure the system in response to faults and availability requests, and the Ensemble protocol stack to provide group communication services. Furthermore, AQuA uses gateways to provide fault-tolerant mechanisms and group communication services to standard CORBA applications transparently as directed by the QuO runtime and the Proteus manager.
The AQuA architecture thus raises the level of abstraction at which a programmer thinks about fault tolerance much higher, relative to existing group communication systems, allowing an application programmer high-level control over the type of faults that should be tolerated and the level of availability desired from a distributed object. In addition, AQuA dynamically adapts the configuration of a system at runtime, in response to faults that occur, to try to maintain a desired level of availability. Finally, it recognizes that applications will require different levels of availability during different phases of their executions, and supports system reconfiguration in response to changing application requirements.