Extended QoS modelling based on multi-application environment in network on chip

Abstract

Until now, there is no standard method of the quality of service (QoS) measurement and fewer techniques have been used to provide its definition. Therefore, researchers are looking for a projection of QoS on quantifiable space, since it is qualitative, subjective and not measurable. However, a few tentatives have studied QoS parameter estimation. Many applications in network on chip (NoC) present variable QoS parameters such as packet loss rate (PLR), end-to-end delay (EED) and throughput (Thp). However, there are a few papers that have developed different methods to modelise QoS in NoC. Their QoS presentation does not provide a multi-application parameter arbiter. Independently of the approach used, an important challenge associated with QoS provision is the development of an efficient and flexible way to monitor QoS. The originality of our approach is based on a proposition of a QoS–intellectual property module in NoC architecture to improve network performances. We implement an extended approach of QoS metrics modelling for NoC on multi-parameter and multi-application environment. The QoS metrics model is based on QoS parameters such as PLR, EED and Thp for different applications. To validate this work, a dynamic routing simulation for 4 × 4 mesh NoC behaviour under three different applications, namely transmission control protocol, variable bit rate and constant bit rate, is considered. To achieve an ideal network behaviour, load balancing on NoC with multiple concurrent applications is improved using QoS metrics measurement based on dynamic routing. The results have shown that extended QoS modelling approach is easy and cheap to implement in hardware–software quantifiable representation. Thus, implementing a quantifiable representation of QoS can be used to provide a NoC services arbiter. QoS arbiter interacts with other routers to ensure flit flow and QoS modelling to provide a QoS value.