Fault Tolerance and Reliability of Mesh Multi-computer Networks – A Review

Vol.4 No.3

Year: 2010

Issue: January - March

Title: Fault Tolerance and Reliability of Mesh Multi-computer Networks – A Review   

Author Name: Mostafa Abd-el-barr   

Synopsis:   

Multi-computer systems (MCSs) are efficient in solving computing-intensive problems. A MCS consists of a number of processing elements (PEs) and an interconnection network (IN). A fault in any PE and/or the IN can lead to losses in sensitive data and/or the overall throughput. In this paper, we provide a review of the fault tolerance and reliability assessing techniques of mesh MCSs. A number of fault-tolerant routing techniques are analyzed including the dimension ordering, the turn model, and the block-fault model. In reliability analysis, we consider the sub-mesh reliability exact and approximate models and the task-based reliability computation.  It is expected that some of the techniques and algorithms covered in this paper can have applications in the domain of wireless mesh networks.