TelegraphCQ: Continuous Dataflow Processing for an Uncertain World
Dr. Sailesh Krishnamurthy
University of California, Berkeley.
Increasingly pervasive networks are leading towards a world where data is constantly in motion. In such a world, conventional techniques for query processing, which were developed under the assumption of a far more static and predictable computational environment, will not be sufficient. In response to this need, the Telegraph project at Berkeley has developed a suite of novel technologies for continuously adaptive query processing. We are currently building the next generation Telegraph system, called TelegraphCQ, which is focused on meeting the challenges that arise in handling large numbers of continuous queries over high-volume, highly-variable data streams. In this talk, I will describe the TelegraphCQ system architecture and its underlying technology, and report on our ongoing implementation effort leveraging the PostgreSQL open source code base.
In addition, I will present a demonstration of sharing and adaptivity in the current TelegraphCQ system (including live data if it's feasible to get connected to the internet)