Classification of Traffic Profiles in On-chip Networks

The development and design of networks-on-chip (NOCs) is an important scientific and technical task, and the huge number of studies, articles and publications on this subject proves the increasing attention to this topic and shows its relevance. The interest in the research of traffic profiles in networks-on-chip is due to their influence on the network throughput, which improves its performance. An improperly chosen traffic profile can not only reduce the potency of the system but also fail to interact with individual topologies. Currently there is no detailed research showing the most effective traffic profile selection, this area requires in-depth analysis to optimize network performance. The aim of the study is to provide a detailed review of existing traffic profiles in the NOCs, analyze and classify them, and study and compare different types of traffic using high-level network-on-chip simulators to analyze their applicability in NOCs of different configurations depending on the network topology. The results of this work will help to identify the most effective combinations of traffic types and topologies to improve the performance of networks-on-chip.