Analysing the cloud
A vast amount of information is stored in online cloud systems, which may mean competitive advantage for any business able to analyse it. A lack of suitable software currently makes such analysis difficult, thus Europe may benefit from investing in such technologies. The EU-funded 'Business intelligence for the masses' (BI4MASSES) project aimed to design and implement scalable cloud analytics software. The undertaking consisted of four phases: introduction of a distributed cloud infrastructure service, data mining and reporting, real-time stream processing and development of applications. The project ran between mid-2010 and mid-2014. Team members first designed and implemented a suitable cloud infrastructure onto which several distributed systems data processing tools were installed. Next came design and implementation of a data mining and reporting service, using open-source tools. Although the work led to several publications, the prototypes could not be expanded to support millions of users as intended. The reason was lack of a sustainable business model, which the project's Principal Investigator and one student overcame by establishing a separate company. Subsequently, the project developed a prototype real-time data stream and complex event processing service. Lastly, the project created and demonstrated intelligent applications based on the previously developed platform services. The applications were delivered to a variety of large companies; the tools were applied to analysing various kinds of networking logs and validating sensor data. The project yielded five Masters theses. Three PhD students were continuing studies at the time of the project's closure. BI4MASSES yielded new software tools for analysing cloud data for business applications.
Keywords
Cloud, network analysis, sensor validation, business intelligence, data mining
