DRAFT VERSION 1 A comparison of ABR and UBR to support TCPtra

1 A comparison of ABR and UBR to support TCP tra c Sam Manthorpe and Jean-Yves Le Boudec Abstract| This paper compares the performance of ABR and UBR for providing high-speed network interconnection services for TCP tra c. We test the hypothesis that UBR with adequate bu ering in the ATM switches results in better overall goodput for TCP tra c than explicit rate ABR for LAN interconnection. This is shown to be true in a wide selection of scenarios. Four phenomena that may lead to bad ABR performance are identi ed and we test whether each of these has a signi cant impact on TCP goodput. This reveals that the extra delay incurred in the ABR end-systems and the overhead of RM cells account for the di erence in ABR and UBR performance. We test whether it is better to use ABR to push congestion to the end-systems in a parking-lot scenario or whether we can allow congestion to occur in the network. Finally, we test whether the presence of a `multiplexing loop' causes performance degradation for ABR and UBR. We nd our original hypothesis to be true in all cases. We observe, however, that ABR is able to improve performance when the bu ering inside the ABR part of the network is small compared to that available at the ABR end-systems. We also see that ABR allows the network to control fairness between end-systems. Keywords|TCP, ABR, UBR, EPD, performance