Kaiwen Zhang, Postdoctoral Fellow in Computer Science at the Technical University of Munich
Towards Big Data Velocity in Event-based Systems
Velocity is the next challenge in Big Data, since the value obtained from analyzing the data is highly dependent on time: real-time analytics must be performed over freshly generated data in order to extract actionable insights. To accelerate real-time analytics, Big Data systems must be capable of processing "data in motion": data created continuously which is analyzed while being disseminated from its source and before it is stored in a database.
In this talk, we first present our past works which adopt this notion of data in motion in the context of event-based systems, particularly, publish/subscribe systems. Our research achieve velocity at a large scale by aggressively processing data within the dissemination system, thus reducing the volume of traffic to be carried in the network. We then provide an outlook of ongoing and future works, which integrate event processing and dissemination with other systems and technologies, such as Hbase and OpenFlow.
Kaiwen Zhang is an Alexander von Humboldt postdoctoral fellow in Computer Science at the TU Munich since 2015 and a member of the Middleware Systems Research Group since 2010. Born in Beijing (China), he obtained his B.Sc and M.Sc at McGill University in Montréal and his Ph.D at the University of Toronto. His research interests include event-based systems, real-time Big Data systems, massively multiplayer online games, and software-defined networking.