Comparison of Packet Switch Architectures and Pacing Algorithms for Very Small Optical RAM

Abstract—One of the difficulties with optical packet switched (OPS) networks is buffering optical packets in the network. The research on optical RAM presently being done is not expected to achieve a large capacity soon. However, the burstiness of Internet traffic causes high packet drop rates and low utilization in small buffered OPS networks. In this article, we investigate and compare optical-buffered switch architectures and pacing algorithms for minimizing the buffer requirements of OPS switches. We simulate two mesh topologies (NSFNET and Abilene) for goodput and packet drop rate comparisons and optimization of XCP parameters. We show that XCP-based pacing algorithm with a shared buffered switch architecture yields high TCP goodput and low packet drop rate in a core OPS network when very small optical RAM buffers are used. Keywords-small buffer, OPS, optical RAM, optical switch I