This paper presents a new analytical model for the estimation of the performance of TCP connections. The model is based on the description of the behavior of TCP-Tahoe in terms of a closed queueing network, whose solution can be obtained with very low cost, even when the number of TCP connections that interact over the underlying IP network is huge. The protocol model can be very accurate, deriving directly from the finite state machine description of the protocol. The assessment of the accuracy of the analytical model is based on comparisons against detailed simulation experiments developed with the ns-2 package. Numerical results indicate that the proposed closed queueing network model provides extremely accurate performance estimates, not only for average values, but even for distributions, in the case of the classical single-bottleneck configuration, as well as in more complex networking setups
A Detailed and Accurate Closed Queueing Network Model of Many Interacting TCP Flows
GARETTO, MICHELE;
2001-01-01
Abstract
This paper presents a new analytical model for the estimation of the performance of TCP connections. The model is based on the description of the behavior of TCP-Tahoe in terms of a closed queueing network, whose solution can be obtained with very low cost, even when the number of TCP connections that interact over the underlying IP network is huge. The protocol model can be very accurate, deriving directly from the finite state machine description of the protocol. The assessment of the accuracy of the analytical model is based on comparisons against detailed simulation experiments developed with the ns-2 package. Numerical results indicate that the proposed closed queueing network model provides extremely accurate performance estimates, not only for average values, but even for distributions, in the case of the classical single-bottleneck configuration, as well as in more complex networking setupsI documenti in IRIS sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.