Dependable services in distributed systems rely on some kind of agreement. Such an agreement can be obtained by solving the consensus problem. Most of the proposed consensus' algorithms are based on mutual knowledge of the participants and thus inadequate to wide area networks (WANs). In previous papers, we proposed protocols which deal with WANs constituted of interconnected physical groups of machines (LANs/domains). These protocols rely on Chandra and Toueg' unreliable failure detectors model for asynchronous systems extended to handle broadcast addresses and safeness of a whole subnet. Nevertheless, we assumed the same characteristics for local and distant failure detectors. In the present paper, we propose a new algorithm which takes into account a different behavior for the detectors. More precisely, local detectors behave as the S class of Chandra and Toueg and so are more reliable than the distant detectors which behave as the \PiS class. Moreover, the distant failure detector we dene, has the ability to test the failure of a local network via a broadcast address. We prove the correctness of the new algorithm and give some implementations hints in the Internet context.

This paper reports ongoing research on the design and implementation of a CORBA [11, 12] object replication service (ORS). The absence of such a service in the actual CORBA specication de- nitely increases the diOEculty to implement and to port dependable applications, namely in the eld of distributed database systems, and consequently may compromise the suitability of the standard. Two major guidelines have been adopted. First, the ORS should be reliable (i.e., preserve strong replica consistency) without a priori constraints on the underlying distributed system. In other words, replica consistency should be preserved regardless of the level of synchronism and type of failures of the system. Second, the whole design and implementation of the ORS should comply with the OMG CORBA 2.0 specication.

Despite the great number of papers that broach the consensus agreement, just a few specifically discuss the structure or the implementation of failure detectors. Even these papers usually present only the structure of the detectors, and avoid presenting a comparison between the suggested detector and other existing ones. In fact, at the literature, we have found only one work that compares by simulations different failure detector models. We have chosen to use that paper as basis, and decided to make a comparison among some failure detectors models, using practical systems. In a practical environment, the implementation issues and the operational system can influence both the performance of the detectors and the consensus termination time. This paper presents our results, which allows us not only to judge the efficiency of these models, but also allow us to learn how these detectors work under different failure situations and under diverse system loads.