James E. Allchin; Martin S. McKendry

techreport

We introduce an approach to robust computation, in distributed systems. This approach is the foundation of reliablity in the Clouds decentralized operating system. it is based on atomic actions operating on instances of abstract data types(objects). We present an event-based model of computation in which scheduling of responses to operation invocations is controlled by objects. We discuss an integrated strategy for synchronization and recovery which uses relationships betweenthe abstract states of objectss to track dependencies between actions. Serilizability is defined in terms of the semantics of operations. This permits high concurrency to be obtained in non-serializable implementations without deviation from serializable abstract behavior. We define a class of schedulers that allows objecs to make autonomous scheduling decisions. We present the use of non-serializable operation semantics. Finally we discuss implementation of the model, includind action synchronization, object operation odering using action-based counting semaphores, and action recovery.

Allchin

Georgia Insititute of Technology

2001-01-01

1. Introduction 2. Background 3. An event-based Computation Model 4. Abstract State/ Dependency Model 5. Non-Serializble Histories 6. Implementing the Model 7. Summary 8. Acknowledgements 9. References

A hard-copy of this is in the Papers Cabinet

Synchronization and Recovery of Actions