Fractional lambda switching

Fractional lambda switching (FλS) leverages on Time-driven switching (TDS) to realize sub-lambda switching in highly scalable dynamic optical networking , which requires minimum (possibly optical) buffers. In this context, TDS has the same general objectives as optical burst switching and optical packet switching: realizing all-optical networks with high wavelength utilization. TFs can be viewed as virtual containers for multiple IP packets that are switched at every TDS switch based on and coordinated by the UTC signal. In the context of optical networks, SVPs are called fractional lambda pipes (FλPs). Since non-immediate forwarding requires buffering packets in network switches, with current optical technologies its extensive deployment in all-optical FλS switches would be expensive. However, limited non-immediate forwarding capability with support for small forwarding delays, i.e., small values of Dp, could be included in switches to decrease blocking probability. Although this will require optical buffering, its requirements in terms of size and access are far simpler than the one posed by optical bust switching and optical packet switching. In TDS all packets in the same TF are switched in the same way. Consequently, header processing is not required, which results in low complexity (hence high scalability) and enables optical implementation. The TF is the basic SVP capacity allocation unit; hence, the allocation granularity depends on the number of TFs per time cycle. For example, with a 10 Gb/s optical channel and 1000 TFs in each time cycle, the minimum FλP capacity (obtained by allocating one TF in every time cycle) is 10 Mb/s. Scheduling through a switching fabric is based on a pre-defined schedule, which enables the implementation of a simple controller. Moreover, low-complexity switching fabric architectures, such as Banyan, can be deployed notwithstanding their blocking features, thus further enhancing scalability. In fact, blocking can be avoided during schedule computation by avoiding conflicting input/output connections during the same TF. Previous results [4] show that (especially if multiple wavelength division multiplexing channels are deployed on optical links between fractional λ switches) high link utilization can be achieved with negligible blocking using a Banyan network without speedup.