Abstract: "Currently existing message logging protocols demonstrate a classic pessimistic vs. optimistic tradeoff. We show that the optimistic-pessimistic tradeoff is not inherent to the problem of message logging. We construct a message-logging protocol that has the positive features of both optimistic and pessimistic protocol: our protocol prevents orphans and allows simple failure recovery; however, it requires no blocking in failure-free runs. Furthermore, this protocol does not introduce any additional message overhead as compared to one implemented for a system in which messages may be lost but processes do not crash."

Concurrency in Dependable Computing focuses on concurrency related issues in the area of dependable computing. Failures of system components, be hardware units or software modules, can be viewed as undesirable events occurring concurrently with a set of normal system events. Achieving dependability therefore is closely related to, and also benefits from, concurrency theory and formalisms. This beneficial relationship appears to manifest into three strands of work. Application level structuring of concurrent activities. Concepts such as atomic actions, conversations, exception handling, view synchrony, etc., are useful in structuring concurrent activities so as to facilitate attempts at coping with the effects of component failures. Replication induced concurrency management. Replication is a widely used technique for achieving reliability. Replica management essentially involves ensuring that replicas perceive concurrent events identically. Application of concurrency formalisms for dependability assurance. Fault-tolerant algorithms are harder to verify than their fault-free counterparts due to the fact that the impact of component faults at each state need to be considered in addition to valid state transitions. CSP, Petri nets, CCS are useful tools to specify and verify fault-tolerant designs and protocols. Concurrency in Dependable Computing explores many significant issues in all three strands. To this end, it is composed as a collection of papers written by authors well-known in their respective areas of research. To ensure quality, the papers are reviewed by a panel of at least three experts in the relevant area.

Middleware is everywhere. Ever since the advent of sockets and other virtu- circuit abstractions, researchers have been looking for ways to incorporate high- value concepts into distributed systems platforms. Most distributed applications, especially Internet applications, are now programmed using such middleware platforms. Prior to 1998, there were several major conferences and workshops at which research into middleware was reported, including ICODP (International C- ference on Open Distributed Processing), ICDP (International Conference on Distributed Platforms) and SDNE (Services in Distributed and Networked - vironments). Middleware’98was a synthesis of these three conferences. Middleware 2000 continued the excellent tradition of Middleware’98. It p- vided a single venue for reporting state-of-the-art results in the provision of distributed systems platforms. The focus of Middleware 2000 was the design, implementation, deployment, and evaluation of distributed systems platforms and architectures for future networked environments. Among the 70 initial submissions to Middleware 2000, 21 papers were - lected for inclusion in the technical program of the conference. Every paper was reviewed by four members of the program committee. The papers were judged - cording to their originality, presentation quality, and relevance to the conference topics. The accepted papers cover various subjects such as caching, re?ection, quality of service, and transactions.

This book constitutes the refereed proceedings of the First International Conference on High-Performance Computing and Communications, HPCC 2005, held in Sorrento, Italy in September 2005. The 76 revised full papers and 44 revised short papers presented were carefully reviewed and selected from 273 submissions. The papers are organized in topical sections on network protocols, routing, and algorithms; languages and compilers for HPC; parallel and distributed system architectures; embedded systems; parallel and distributed algorithms, wireless and mobile computing, Web services and Internet computing; peer-to-peer computing, grid and cluster computing, reliability, fault-tolerance, and security; performance evaluation and measurement; tools and environments for software development; distributed systems and applications; high performance scientific and engineering computing; database applications and data mining; HPSRF; pervasive computing and communications; and LMS.

The primary audience for this book are advanced undergraduate students and graduate students. Computer architecture, as it happened in other fields such as electronics, evolved from the small to the large, that is, it left the realm of low-level hardware constructs, and gained new dimensions, as distributed systems became the keyword for system implementation. As such, the system architect, today, assembles pieces of hardware that are at least as large as a computer or a network router or a LAN hub, and assigns pieces of software that are self-contained, such as client or server programs, Java applets or pro tocol modules, to those hardware components. The freedom she/he now has, is tremendously challenging. The problems alas, have increased too. What was before mastered and tested carefully before a fully-fledged mainframe or a closely-coupled computer cluster came out on the market, is today left to the responsibility of computer engineers and scientists invested in the role of system architects, who fulfil this role on behalf of software vendors and in tegrators, add-value system developers, R&D institutes, and final users. As system complexity, size and diversity grow, so increases the probability of in consistency, unreliability, non responsiveness and insecurity, not to mention the management overhead. What System Architects Need to Know The insight such an architect must have includes but goes well beyond, the functional properties of distributed systems.