Design of multi-path data routing algorithm based on network reliability

Data routing through an interconnected network is important and this paper addresses the design of a multi-path data routing algorithm based on network reliability. Generally, multiple routes for a given source–terminal pair exist in a data network and the best possible route based on network metrics like hop-count, delay, traffic, queue, etc. is selected by a routing algorithm. Since network reliability incorporates all these metrics, the routing decision based on reliability seems to be the best possible option and a distributed routing algorithm based on the source–terminal (s–t) path reliability has been proposed in this work. Each node in the proposed routing generates an adjacency matrix of a network graph by periodically exchanging connection information with the adjacent nodes and selects multiple routes based on reliability of the paths. We propose an implementation of a two-path routing algorithm that instead of one includes two next-hop nodes in each node’s routing table. An example is given for further illustration of the proposed algorithm.

Graphical abstractFigure optionsDownload full-size imageDownload as PowerPoint slideHighlights► The present work proposes an efficient path-reliability-based routing technique. ► It analyses all possible paths and selects the best reliable path(s). ► The inclusion of more than one next-node towards a same destination is proposed. ► At least two-path routing could be implemented considering two best possible paths. ► The path-based routing is scalable, avoids packets circulation and count-to-infinity problem.