Local area network characteristics, with implications for broadband network congestion management

Abstract

This paper examines the phenomenon of congestion in order to better understand the congestion management techniques that will be needed in high-speed, cell-based networks. The first step of this study is to use high time-resolution local area network (LAN) traffic data to explore the nature of LAN traffic variability. Then, we use the data for a trace-driven simulation of a connectionless service that provides LAN interconnection. The simulation allows us to characterize what congestion might look like in a high-speed, cell-based network. The most striking aspect of the LAN data is the extreme traffic variability on time scales ranging from milliseconds to months. Conventional traffic models do not capture this behavior, which has a profound impact on the nature of traffic congestion. When our realistic data is applied to simple models of LAN interconnection, we observe that: During congested periods, congestion persists and losses can be significant. Congestion losses cannot be avoided by modest increases in buffer capacity. The consequences of misengineering can be serious. Fortunately, most congested periods are preceded by signs of impending danger.

About Network Traffic Models

“…Some studies show that the time interval between packages is not explicitly distributed. For example, the studies described in [5][6][7][8] confirm that the distribution of the time interval between the packet entries in the local and global networks differs from the exponential distribution. Therefore, the statistical "self-similar" processes, which differ from the Poisson process for their network traffic modeling, began to be used [9,10].…”

Approximating sliding windows by cyclic tree-like histograms for efficient range queries

“…The data distribution is depicted in Figure 6. This real-life data set has been used also in [51,66,57,36]. The statistics of the most significative data sets are reported in Table 4.…”

Multihop Virtual Topology Design in WDM Optical Networks for Self-Similar Traffic

“…Network arrivals are often modeled as Poisson processes for analytic simplicity. However, a number of studies with the Internet traffic has shown that both local-area and wide-area network inter-arrival packet distribution clearly differs from exponential [18][19][20]. Leland et al [21], in their paper, showed that the LAN traffic is better modeled using statistically Self-Similar processes.…”