Traffic performance of buffering strategies in all-optical nodes for IP networks

In order to support the continuous growth of transmission capacity demand, optical packet switching technology is emerging as a strong candidate, promising to allow fast dynamic allocation of WDM channels, combined with a high degree of statistical resource sharing. This work addresses the comparison of input and shared buffering in an optical switch architecture based on an arrayed waveguide grating to route packets. Since a limited maximum buffer depth is feasible nowadays, particular care must be paid to buffer implementation. The performance evaluation considers different length distributions of packets offered to the node.