Extensible Context-aware Stream Processing on the Cloud: Rationale and Challenges

Walid Aref, Purdue University

NSF Workshop on Social Networks and Mobility in the Cloud, February 2012, Arlington, VA


Rationale and Challenges for Massive Data Stream Processing on the Cloud
The ubiquity of mobile devices, location services, and sensor pervasiveness, e.g., as in smart city initiatives, call for scalable computing platforms and massively parallel architectures to process the vast amounts of the generated streamed data. Cloud computing provides some of the features needed for these massive data streaming applications. For example, the dynamic allocation of resources on an as-needed basis addresses the variability in sensor and location data distributions over time. However, today’s cloud computing platforms lack very important features that are necessary in order to support the massive amounts of data streams envisioned by the massive and ubiquitous dissemination of sensors and mobile devices of all sorts in smart-city-scale applications.