A measurement analysis of ATM cell-level aggregate traffic

Advances in network technologies have out-paced our abilities to effectively manage and engineer them. A fundamental challenge in traffic management of high-speed networks is to describe, analyze and control complex traffic flows in ways that can be applied in practice. Our approach is experimental in nature and starts with collection and analysis of high-resolution traffic traces from working high-speed networks. This paper reports on an analysis of busy-hour cell-level ATM traffic measurements collected from a commercial ATM network operated by a major North American carrier through a custom-built ATM traffic recording. We present results on ATM cell-level analysis with focus on the aggregate traffic level. Traffic is dominated by a few VPI/VCIs which is not uncommon in emerging network environments. The aggregate traffic (without the dominating VPI/VCI that contributes more than 80% of the traffic) appears to be self-similar, providing evidence that aggregate ATM traffic can have long-range dependent features. This is consistent with findings from lower speed packet networks.