In [5] we proposed the Load Balanced WASPNET Scheduler (LBWS), a novel scheduling approach for WASPNET optical packet switching (OPS) nodes. Unlike the original WASPNET description, LBWS is fully distributed and its computational cost is independent from switch size. Since it predicts packet delays at packet ingress based on a deterministic evolution of the switch configuration, it is not necessary to emulate RAM memories. However, LBWS does not maintain the packet sequence. In this paper we present the LBWS algorithm with packet ordering (OLBWS), the first step to prevent mis-sequencing in a LBWS scheduler. We show that despite the algorithm simplicity a reasonable throughput can be achieved.
展开▼
