Communication paradigms in distributed systems infrastructure
Mohamed Salem Elshebani · 2009
There has recently been increasing prominence of distributed systems, to gain access to distributed services around the globe, without a real concern for the underlying system environment. Therefore, after the Internet explosion, there are massive number of standards and platforms, and trial for convergence among different distributed systems architectures, became a real task. However, the main concern of this paper, is to undertake the topic of communication paradigms, in terms of system infrastructure, and reliability. This work is also dealing with distributed communications, which describe how processors at different sites, can communicate, to reach to a system with fault- tolerance capability. However, Distributed System (D.S) should be fault tolerant, and in this paper, a method is adopted to handle fault detection, reconfiguration, and recovery from failure. In terms of Algorithms and theorems directly related to these criteria, to achieve fault tolerant system. Therefore, Algorithm 1.0, is related to fault detection scheme, by applying a timeout with acknowledgements. Algorithm 2.0, is describing the reconfiguration, as a basic and reliable broadcast. Algorithm 3.0, is related to recovery from failure, as a handshaking algorithm. Algorithm 4.0 is related to fault tolerant scheme, in which consensus can be reached if crash occurs. Theorems are also introduced in this work, as complement to algorithms laid down, in theorem 1.0, the claims were about the system robustness criteria, whereas, Theorem 2.0 is dealing with handshaking procedure, and system recovery. Theorem 3.0 is presented to address the consensus problem which is solved by algorithm3.0. In this paper, a viewpoint analogy is also considered in terms of distributed system consolidation, for different platforms, to ensure the openness of the same system.