It is shown that if there is no bound on the number of faulty processes, then in a precise sense, in a system with unreliable but fair communication, Uniform Distributed Coordination (UDC) can be attained if and only if a system has perfect failure detectors.This result is generalized to the case where there is a bound t on the number of faulty processes.It is shown that a certain type of generalized failure detector is necessary and sufficient for achieving UDC in a context with at most t faulty processes.Reasoning about processes' knowledge as to which other processes are faulty plays a key role in the analysis.

This paper describes the design and implementation of a fault-tolerant CORBA naming service-CosNaming FT. Every CORBA object is accessed through its Interoperable Object Reference (IOR), which is registered with the CORBA name service. The name service therefore is a critical gateway to all objects in a distributed system; to avoid having a single point of failure, the name service should be made fault-tolerant. CosNamingFT uses the GroupPac package, a CORBA-compliant suite of protocols, to replicate the name server. GroupPac services are built from Common Object Services that function as building blocks to implement fault-tolerant applications. This paper aims to demonstrate the usefulness of some of these object services and to demonstrate the importance of open solutions issues in replicating distributed objects.

The Nile system is a distributed environment for running very large, data-intensive applications across a network of commodity workstations. These applications process data from elementary particle collisions, generated by the Cornell Electron Storage Ring, and are used by physicists of the CLEO experiment. The applications have a simple data-parallel structure, and so Nile executes them using as much parallelism as is available. Nile currently runs at any single site. It is being used by alpha testers and is scheduled for beta release in March 1998. We describe how we are adapting this local-area Nile system to allow for wide-area, multiple site interactions. In particular, we consider the two problems of scaling and of fault tolerance.

We describe the Sage project, a new approach to software engineering for (fault-tolerant) distributed applications. Sage uses the modal logic of knowledge and applies theoretical results detailing how processes learn facts about each other's state to derive the minimal communication graph for a wide range of coordination problems. The specification interface is controlled, yet expressive enough to capture important distributed coordination problems and weaker variants appropriate for wide-area applications. The resulting graphical display shows programmers which messages must be received. Sage allows users to experiment on the derived protocol by crashing processes, reordering events, losing messages, and partitioning the network. If a solution still exists, Sage regenerates the communication graph. This animates the effects of unpredictable system events on distributed applications, and separates the issues in testing a protocol's behavior in the face of failures, from the effects background system conditions can have on the testing procedure itself.

No abstract available.

No abstract available.

We devetop necessary conditions for the development of asynchronous distributed sofiware that will perform uniform actions (’evenis that if performed by any pro-cess, must be performed at all processes). The pa-per focuses on dynamic uniformity, which differs from ihe classical problems in that processes continually leave and join the ongoing computation. It relates the problem to asynchronous Consensus, and shows that Consensus is a harder problem. We provide a rigorous characterization of the framework upon which several existing distributed programming environments are based. And, our work shows that progress is some-times possible in a primary-partition model even when consensus is not. 1

Agreement on the membership of a group of processes in a distributed system is a basic problem that arises in a wide range of applications. Such groups occur when a set of processes co-operate to perform some task, share memory, monitor one another, subdivide a computation, and so forth. In this paper we discuss the Group Membership Problem as it relates to failure detection in asynchronous, distributed systems. We present a rigorous, formal specification for group membership under this interpretation. We then present a solution for this problem that improves upon previous work.