Data Analytics and the Networks that Enable them
Everyone has more data than they can cope with. Data analytics has some canonical use cases. Cloud computing is creating newer use-cases and is making some of the traditional ones more widely accessible. Mobile computing and context based applications are stressing both the wireless links and the Internet.
We think relatively little has been done in understanding how the underlying networks fare. Sometimes, the network is indeed a pipe -- it is all about the throughput and the latency on offer. Often, more is needed -- predictability, proportional resource allocation, mapping communication patterns to network architectures, private/personal access to data and compute. Architecting for manageability and application agility continues to remains a challenge.
We see two trends: (1) “Big Data” is big, and the need for data analytics is exploding, and (2) networks are becoming increasingly programmable. Understanding the requirements for data analytics can help us design and develop networks that enable applications of the future.
This inter-disciplinary summit, organized by the Mobility & Networking Research Group at MSR, will bring together leading thinkers from academia and industry working in the fields of data analytics, machine learning, networking, and mobility. It will provide a unique opportunity for us to mingle and discuss as we find a path forward.