Alcaudon: Streaming data platform.
Fernández Castaño, Francisco
MetadatosMostrar el registro completo del ítem
Currently, the amount of collected data is growing. Internet companies track all user actions in order to provide better experiences. Science experiments generate colossal amounts of data. Another growing source of data is Internet of Things, i.e. autonomous cars need to collect and analyze large amounts of data in order to function properly. These, combined with the plunging cost of data storage makes storing large amounts of information for later analysis very compelling. Organizations should take advantage of all these data in order to overcome ad-hoc business decisions in favor of data-driven approaches. Therefore, to satisfy data analysis needs, many data processing systems have appeared in the last years. Those systems were focused on processing large amounts of historical data. However, there is an increasing need to acquire knowledge from data as soon as possible, allowing a prompt reaction to a wide range of events. This project aims to implement a distributed streaming data processing platform, Alcaudon, allowing near real-time data analysis of unbounded data-sets. One of the objectives of this project is to provide a powerful programming model, making it straightforward to write distributed data analysis pipelines. Moreover, Alcaudon has been designed contemplating scalability and fault-tolerance. Therefore, this project presents a robust and potent tool to process and analyze unbounded data-sets.