Constrained Computing at the Edge
Clouds and Distributed Computing
Scientific Software Development
TimeMonday, June 25th4pm - 4:20pm
DescriptionBig data and deep learning are the memes of the day, as we shift from a world where data was rare, precious, and expensive to one where it is ubiquitous, commonplace, and inexpensive. Massive digital data (from scientific instruments and IoT devices), powerful multilayer classification networks, and inexpensive hardware accelerators are bringing new data-driven approaches to the edge, challenging long held beliefs about technical computing approaches and illuminating old questions in new ways. How do we best develop workflows that must span both edge devices and cloud or HPC facilities? How do we best meet soft real-time constraints for streaming data, subject to energy, communication, and cost constraints? Drawing on insights from the Big Data and Extreme-scale Computing (BDEC) workshops, this talk will discuss these challenges – how we reached this point --- and where we are likely to go, along with descriptions of technology in place to enable exploration of these challenges.