Flow control and service differentiation in optical burst switching networks
Item Usage Stats
Optical Burst Switching (OBS) is being considered as a candidate architecture for the next generation optical Internet. The central idea behind OBS is the assembly of client packets into longer bursts at the edge of an OBS domain and the promise of optical technologies to enable switch reconfiguration at the burst level therefore providing a near-term optical networking solution with finer switching granularity in the optical domain. In conventional OBS, bursts are injected to the network immediately after their assembly irrespective of the loading on the links, which in turn leads to uncontrolled burst losses and deteriorating performance for end users. Another key concern related to OBS is the difficulty of supporting QoS (Quality of Service) in the optical domain whereas support of differentiated services via per-class queueing is very common in current electronically switched networks. In this thesis, we propose a new control plane protocol, called Differentiated ABR (D-ABR), for flow control (i.e., burst shaping) and service differentiation in optical burst switching networks. Using D-ABR, we show with the aid of simulations that the optical network can be designed to work at any desired burst blocking probability by the flow control service of the proposed architecture. The proposed architecture requires certain modifications to the existing control plane mechanisms as well as incorporation of advanced scheduling mechanisms at the ingress nodes; however we do not make any specific assumptions on the data plane of the optical nodes. With this protocol, it is possible to almost perfectly isolate high priority and low priority traffic throughout the optical network as in the strict priority-based service differentiation in electronically switched networks. Moreover, the proposed architecture moves the congestion away from the OBS domain to the edges of the network where it is possible to employ advanced queueing and buffer management mechanisms. We also conjecture that such a controlled OBS architecture may reduce the number of costly Wavelength Converters (WC) and Fiber Delay Lines (FDL) that are used for contention resolution inside an OBS domain.