Showing papers by "Sonia Fahmy published in 1996"
••
TL;DR: The rules that the sources have to follow to achieve a fair and efficient allocation of network resources are explained.
Abstract: The available bit rate (ABR) service has been developed to support data applications over asynchronous transfer mode (ATM) networks. The network continuously monitors its traffic and provides feedback to the source end systems. This article explains the rules that the sources have to follow to achieve a fair and efficient allocation of network resources.
99 citations
•
19 Jul 1996
TL;DR: In this paper, a congestion avoidance scheme for data traffic in ATM networks is proposed, which has its optimal operating point at 100% utilization and a fixed, non-zero queue delay.
Abstract: A congestion avoidance scheme for data traffic in ATM networks. The scheme achieves both efficiency and fairness, and exhibits a fast transient response. A congestion avoidance scheme for ATM networks is described which has its optimal operating point at 100% utilization and a fixed, non-zero queue delay. The scheme improves control of end-to-end delay and keeps link utilization of expensive links high despite idle periods in the input load.
68 citations
••
18 Nov 1996TL;DR: The paper shows that it is possible to get maximum TCP throughput when there are enough buffers at the switches, however, when the number of buffers is smaller, there is a large reduction in TCP throughput even though the cell loss ratio (CLR) is very small.
Abstract: The available bit rate (ABR) service has been developed to support data applications over asynchronous transfer mode (ATM). It is hence interesting to study the performance of reliable data transport protocols like the "transport control protocol (TCP)" over ABR. We study the effect of running large unidirectional file transfer applications on TCP over ABR with an explicit rate algorithm (ERICA) implemented at the ATM switches. The paper shows that it is possible to get maximum TCP throughput when there are enough buffers at the switches. However, when the number of buffers is smaller, there is a large reduction in TCP throughput even though the cell loss ratio (CLR) is very small. We study the effect of various factors which affect TCP throughput and fairness. These factors include the TCP timer granularity, switch buffering, ABR parameters, and the cell drop policy at the switches.
24 citations
01 Jan 1996