This paper compares the performance of Tail Drop and three different flavors of the RED (Random Early Detection) queue management mechanism: RED with a standard parameter setting, RED with an optimized parameter setting based on a model of RED with TCP flows, and finally a version of RED with a smoother drop function called "gentle RED". Performance is evaluated under various load situations for FTP-like and Web-like flows, respectively. We use measurements and simulations to evaluate the performance of the queue management mechanisms and assess their impact on a set of operator oriented performance metrics. We find that in total (i) no performance improvements of RED compared to Tail Drop can be observed; (ii) fine tuning of RED parameters is not sufficient to cope with undesired RED behavior due to the variability in traffic load; (iii) gentle RED is capable of resolving some of the headaches on RED but not all.

This paper compares the performance of Tail-Drop and three different flavors of the RED (Random Early Detection) queue management mechanism; RED with a standard parameter setting, RED with an optimal parameter setting based on a model of RED with TCP flows, and finally a version of RED with a smooth drop function called gentle RED . Performance is evaluated for FTP-like bulk-data TCP flows and for short-lived Web-like TCP flows in various load situations. In addition to the TCP traffic, the network is lightly loaded with UDP traffic in order to generate a realistic mix of TCP and real-time flows. We use measurements and simulations to evaluate the performance of the queue management mechanisms trying to assess their impact on a set of performance metrics: TCP goodput, UDP loss rate, queuing delay, delay jitter and the distribution of consecutive packet drops.