12.4. Traffic Patterns in E-Business

This section analyzes graphs of traffic patterns derived from actual HTTP logs of a real e-tailer that provided us with sanitized logs. The analysis of patterns of traffic behavior is very useful for predictive purposes. The plots of Figs. 12.2 and 12.3 show different patterns of the daily traffic handled by the e-business site. Figure 12.3 displays some characteristics of customer sessions, defined by a time threshold of thirty minutes. The upper graph of Fig. 12.3 shows the number of active sessions at the beginning of each hour during a typical week. The graph indicates that spikes are smaller than in Fig. 12.2 and that the minimum number of active sessions is around 850. The bottom graph of Fig. 12.3

Get Scaling for E-Business now with the O’Reilly learning platform.

O’Reilly members experience books, live events, courses curated by job role, and more from O’Reilly and nearly 200 top publishers.