No abstract

Since random early detection (RED) was proposed in 1993, many active queue management (AQM) algorithms have been proposed to support better end-to-end Transmission Control Protocol (TCP) congestion control. In this article, the authors introduce and analyze a feedback control model of the TCP/AQM dynamics. Then they suggest the concept of an AQM algorithm that can detect and avoid congestion proactively. Finally, they propose the proportional-integral (PI) proportional-derivative (PD) controller using proportional-integral-derivative (PID) feedback control to overcome the reactive control behavior of existing AQM proposals. The PI-PD controller is able to provide proactive congestion avoidance and control using an adaptive congestion indicator and a control function. A comparative simulation study under a variety of network environments shows that the PI-PD controller outperforms RED and the PI controller in terms of the queue length dynamics, the packet loss rates, and the link utilization.

In this paper, we evaluate the performance of several multicast schemes in optical burst-switched WDM networks taking into accounts the overheads due to control packets and guard bands (GBs) of bursts on separate channels (wavelengths). A straightforward scheme is called Separate Multicasting (S-MCAST) where each source node constructs separate bursts for its multicast (per each multicast session) and unicast traffic. To reduce the overhead due to GBs (and control packets), one may piggyback the multicast traffic in bursts containing unicast traffic using a scheme called Multiple Unicasting (M-UCAST). The third scheme is called Tree-Shared Multicasting (TS-MCAST) whereby multicast traffic belonging to multiple multicast sessions can be mixed together in a burst, which is delivered via a shared multicast tree. In [1], we have evaluated several multicast schemes with static sessions at the flow level. In this paper, we perform a simple analysis for the multicast schemes and evaluate the performance of three multicast schemes, focusing on the case with dynamic sessions in terms of the link utilization, bandwidth consumption, blocking (loss) probability, goodput and the processing loads.