A universal performance factor for multi-criteria evaluation of multistage interconnection networks

Abstract

The choice of an interconnection network for a parallel computer depends on a large number of performance factors which are very often application dependent. We propose a performance evaluation and a comparison methodology. This methodology is applied to a recently introduced class of interconnection networks (Multistage Chordal Ring Based multistage interconnection network). These networks are compared to the well known Omega network of comparable architectural characteristics. The methodology is expected to serve in the evaluation of the use of multistage interconnection networks as an intercommunication medium in today's multi-processor systems.

Quantitative analysis of multilayer switching networks providing multiple internal routes

“…Many indicators designate the behaviour of a fabric, usually operating in a contradictory manner. The Universal performance indicator was introduced in Aljundi et al 30 as a tool for evaluating MINs using a set of metrics and weighting them according to specific needs.…”

Assessment of SDN Controllers in Wireless Environment Using a Multi-Criteria Technique

“…Numerous studies have assessed network behavior by individually examining throughput, latency, and other performance metrics. Aljundi et al [7] and Stergiou et al [8] presented evaluations of multi-stage interconnection networks and multiple input, multiple output (MIMO) systems, respectively. They used typical simulation-based multi-objective methods for network performance evaluation as a time-efficient method able to be implemented given a sufficient quantity of values for each performance factor.…”

Rearranging links: a cost-effective approach to improve the reliability of multistage interconnection networks

“…Cost is an important metric in the design of MINs, so that if a high-performance network have a high cost, then it would not be a suitable case for implementation. According to previous works [14,22,23,25,26,[28][29][30], a common method for estimating the cost of a MIN is calculating the number of crosspoints used in the network. Therefore, the cost is given by the number of crosspoints within a switching element and by the number of switching elements within the network.…”

“…Therefore, we are able to monitor the behavior of the proposed approach in terms of the improvement rate that occurred in the reliability parameter. In addition, like many previous works [14,22,23,25,26,[28][29][30], we will analyze the cost of network hardware to assess the second case. In the cost analysis, it will be assumed that the cost of a switch is proportional to the number of gates involved in it.…”