Throughput analysis of multihop shufflenets in a hot spot traffic scenario: impact of routing strategies

This paper deals with the analysis of the throughput performance of Shuffle Multihop Networks under general traffic configurations and aims at investigating the impact of different classes of routing schemes. A hot-spot traffic scenario was considered, modeling the presence of a centralized network resource to which a quota of the internal traffic is directed or originated from. The analysis was carried out by varying the traffic unbalance degree, from a uniform traffic distribution to a completely unbalanced one (all the traffic is concentrated in the hot spot node). As for the routing algorithms, both minimum path length (fixed or adaptive) and unconstrained path length schemes are studied. Simple analytical expressions of the actual throughput limits were obtained for each class of routing algorithms. By comparing the upper bounds with simulation results, it is shown that there exists an increase in the tightness of the upper bound which can be directly correlated to an increase in the network size. The results obtained show that the throughput performance deteriorates for increasing unbalance of the network load. This worsening can be partially relieved by employing adaptive routing algorithms. These algorithms provide remarkable throughput improvements with respect to fixed routing schemes, but only in the case of moderate hot spot traffic quota. Moreover, results show that unconstrained path length routing schemes lead to the best performance provided that the maximum path length be bounded.