Dual Congestion Awareness scheme in On-Chip Networks

One of the main factors limiting the performance of Networks-on-Chip is congestion. Routing algorithms perform an important role in distributing the traffic load over the network by providing alternative routing paths. The major parts of research works avoid congestion by considering the traffic condition in the forward paths and delivering packets through less congested paths. However, they do not consider the impact of the router arbitration in traffic distribution. On the other hand, traditional methods usually consider the traffic at the node level rather than a region level. In this paper, we propose a method named Dual Congestion Awareness (DuCA) which involves the congestion information about forward paths in the routing decision and backward paths in the router arbitration. DuCA can efficiently use the traffic information to transmit packets through less congested areas (using congestion information about forward paths) and steer traffic from highly congested regions to low congested areas (using congestion information in backward paths). Finally, DuCA is a region-based approach, providing a wider view of the network traffic.