Queuing analysis of shared‐buffer ATM switches with grouped output channels

Abstract

SUMMARYShared bu!ering and channel grouping are powerful techniques with great bene"ts in terms of both performance and implementation. Shared-bu!er switches are known to have better performance and better utilization than input or output queued switches. With channel grouping, a cell is routed to a group of channels instead of a speci"c output channel. In this way, congestion due to output contention can be minimized and the switch performance can therefore be greatly improved. Although each technique is well known by itself in the traditional study of queuing systems, their combined use in ATM networks has not been much explored previously. In this paper, we develop an analytical model for a shared-bu!er ATM switch with grouped output channels. The model is then used to study the switch performance in terms of cell loss probability, cell delay and throughput. In particular, we study the impact of the channel grouping factor on the bu!er requirements. Our results show that grouping the output channels in a shared-bu!er ATM switch leads to considerable savings in bu!er space.

An approximate analysis of shared‐buffer channel‐grouped ATM switches under imbalanced traffic

“…The queuing system and assumptions, as well as the state space, are the same as those described for the uniform traffic in Reference [1]. As stated earlier, only the input traffic statistics are affected by the non-uniformity of the traffic.…”

“…The behaviour of the switching system is characterized by the probabilities of state transitions, which involve cell departure and cell arrival processes. As explained in Reference [1], the state equations for the switch are given as follows: …”

“…With channel grouping, in which a cell is routed to a group of channels instead of a specific output channel, congestion due to output contention can be minimized and the switch performance can therefore be greatly improved. Using these techniques in ATM switches, which are referred to as shared-buffer channelgrouped (SBCG) switches, has shown to yield significant saving in buffer space and better throughput under uniform traffic [1,2]. However, uniform traffic does not represent a realistic view of traffic patterns in real systems, because data that is generated by real applications, such as audio or video transfer, might not utilize the maximum bandwidth.…”

“…Our main goals are to show how the analytical model developed in Reference [1] can be extended to include imbalanced traffic, such as bursty and hot-spot traffic and to investigate the impact of the channel-grouping factor and other key switch design parameters on the switch performance under such traffic conditions. The rest of the paper is organized as follows.…”

“…There has been much interest by the industrial and research communities in devising ATM switches capable of accommodating a variety of traffic sources with conflicting quality of service requirements. The combined use of techniques such as shared buffering and channel grouping in ATM switches has shown great benefits in terms of both performance and implementation [1][2][3][4]. Shared-buffer switches are known to have better performance and better utilization than input or output queued switches.…”