FAIR QUEUEING ALGORITHM Nagle's algorithm, by changing the way packets from different sources interact, does not reward, nor leave others vulnerable to, antisocial behavior. On the surface, this proposal appears to have considerable merit, but we are not aware of any published data on the performance of datagram networks with such fair queueing gateways. In this paper, we will first describe a modification of Nagle's algorithm, and then provide simulation data comparing networks with FQ gateways and those wi The three different components of congestion control algorithms introduced above source flow control, gateway routing and gateway queueing algorithms, interact in interesting and complicated ways. It is impossible to assess the effectiveness of any algorithm without reference to the other components of congestion control in operation We will evaluate our proposed queueing algorithm in the context of static routing and several widely used flow control algorithms. The aim is to find a queueing algorithm that functions well in current computing environments. The algorithm might, indeed it should enable new and improved routing and flow control algorithms, but it must not require We had three goals in writing this paper. The first was to describe a new fair queueing algorithm. In the next section, we discuss the design requirements for an effective ueueing algorithm, outline how Nagle's original proposal fails to meet them, and then propose a new fair queueing algorithm which does meet these requirements. Our second goal was to provide some rigorous understanding of the performance of this new fair queueing algorithm; we present a delay-throughput curve given by this algorithm for a specific configuration of sources, and then compare this performance to that given by the FCFS algorithm. Our third goal was to evaluate our new queueing proposal in the context of real networks. To this end we discuss some currently implemented flow control algorithms and present simulation data comparing several combinations of flow control and queueing algorithms on six benchmark networks In circuit-switched networks where there is explicit buffer reservation and uniform packet sizes, it has been established that round-robin service disciplines allocate bandwidth fairly(Hahne, 1986; Katevenis, 1987). Recently Morgan (1989)has examined the role such queueing algorithms play in controlling congestion in circuit switched networks although his application context is quite different from ours, his conclusions are qualitatively similar. In other related work, the datAKIt queueing algorithm combines round-robin service and FIFO Priority service, and has been analyzed extensively (Lo 1987; Fraser and Morgan, 1984). Also, Luan and Lucantoni(1988) present a different form of bandwidth management policy for circuit switched networks Since the completion of this work, we have learned of similar work by Zhang(1989); her Virtual Clock gateway queueing algorithm is essentially identical to the fair queueing agorithm presented here. Zhang analyzes this algorithm in the context of a proposed resource reservation scheme. the flow Network, whereas we do not consider resource reservation. Heybey and Davin(1989)have simulated a simplified version of our fair queueing algorithm, investigating issues of buffer allocation and policy-based bandwidth allocation. McKenney(1990)and Keshav(1990) have investigated the implementation pects of fair queueing. In addition, Greenberg and Madras(1990)have established some performance bounds on the fairness of our fair queueing scheme and other similar DATAKIT is a Trademark of AT&T