“School of Computer Science”
Back to Papers HomeBack to Papers of School of Computer Science
| Paper IPM / Computer Science / 10903 |
|
||||||||
| Abstract: | |||||||||
|
There exist a large number of graph optimization problems in the literature, which arise in network design and analysis. Our objective in this paper is to highlight the disconnection probability which can arise in interconnect networks of large-scale parallel processors systems. Although traditional measures of fault-tolerance such as reliability and availability are applicable to such systems, these measures were designed mostly for mission-oriented applications or repairable systems. They fail to account for the high redundancy levels typical in peer-to-peer (P2P) communication and distributed systems. For these systems, new measures have been introduced that can evaluate the capability of a system for graceful degradation. In the design of such systems, one of the most fundamental considerations is the reliability of their interconnected networks, which can be usually characterized by connectivity of the topological structure of the network. In this paper, we analyze the problem of network disconnection in the context of large-scale P2P networks and understand how static patterns of node failure affect the resilience of such networks. Simulation results based on the network topology confirm the validity of the analytical approximation and demonstrate the localizer efficiency.
Download TeX format |
|||||||||
| back to top | |||||||||
