Advanced Content Delivery, Streaming, and Cloud Services

Chapter 14Mining Distributed Data Streams on Content Delivery Networks

Eugenio Cesario¹, Carlo Mastroianni¹, and Domenico Talia^1,2

¹ICAR-CNR, Rende (CS), Italy

²DIMES, University of Calabria, Rende (CS), Italy

14.1 Introduction

Mining data streams (DSs) is a very important research topic and has recently attracted a lot of attention, because in many cases data is generated by external sources so rapidly that it may become impossible to store it and analyze it offline. Moreover, in some cases streams of data must be analyzed in real time to provide information about trends, outlier values or regularities that must be signaled as soon as possible. Important application fields for stream mining are as diverse as financial applications, network monitoring, security problems, telecommunication networks, Web applications, content delivery networks, sensor networks, analysis of atmospheric data, and so on.

The mining DS process becomes more difficult when streams are distributed, because mining models must be derived not only for the data of a single stream, but for the integration of multiple and heterogeneous DSs. This scenario can occur in all the application domains mentioned before and specifically in a content delivery network. In this context, user requests delivered to a Web system can be forwarded to any of several servers located in different and possibly distant places, in order to serve requests more efficiently and balance the load among the servers. The analysis of user ...