Performance Evaluation of a MapReduce Hadoop-based Implementation for Processing Large Virtual Campus Log Files

Cloud computing technologies are bringing new scales of computational processing power and storage capacity to meet very demanding requirements of today's applications. One such family of applications is the one of analytics based on processing big data. More specifically, there is a large family of analytics applications from processing log data files. Indeed, log data files are commonplace in many Internet-based systems and applications, comprising system logs, server logs, application logs, databases logs, user activity logs, etc. These applications are analytics oriented applications based on processing the various types of log files. While log data file processing has been recently an issue of investigation by many researchers and developers, the new feature is that of scale: Cloud based systems can enable processing unlimited amount of data either off-line or online in streaming mode. In this work we evaluate the performance of a MapReduce Hadoop-based implementation for processing large log data files of a Virtual Campus. The study aims to reveal the potential of using such implementations as a basis for learning analytics for use by a variety of users in a Virtual Campus.