Controlling High-Bandwidth Flows at the Congested Router

FIFO queueing is simple but does not protect traffic from high-bandwidth flows, which include not only flows that fail to use end-to-end congestion control, but also short round-trip time TCP flows. At the other extreme, per-flow scheduling mechanisms provide max-min fairness but are more complex, keeping state for all flows going through the router. This paper presents RED-PD, a mechanism that combines simplicity and protection by keeping state for just the high-bandwidth flows. RED-PD uses the packet drop history at the router to detect high-bandwidth flows in times of congestion and preferentially drops packets from these flows. This paper discusses the design decisions underlying RED-PD. We show that it is effective at controlling highbandwidth flows using a small amount of state and very simple fast-path operations.