COMPUTATIONAL RESEARCH in BOSTON and BEYOND (CRIBB)

Date	Apr. 4, 2014
Speaker	Tom Graham and Todd Mostak (Co-Founders of Map-D)
Topic	Map-D: A GPU Database for Interactive Big Data Analytics
Abstract:	Interactive big data analytics: Using GPUs to power the world's fastest database. As part of an emerging conversation between HPC and enterprise, this talk will focus on the future of high performance big data analytics from enterprise, government and scientific perspectives while tracking the challenges posed by data collection, hardware integration and interface design. But there is more at stake than data-drive cost savings, these perspectives are framed by the need to socialize and democratize high-power big data analytics to the advantage of all. Map-D is an ultra-fast GPU database that allows anyone to interactively analyze and visualize big data in real time. Built into GPU memory, Map-D's unique architecture runs 70-1000X faster than other in-memory databases and big data analytics platforms. We will also showcase Map-D's public demos, including TweetMap that maps over 1 billion tweets in real time and CampaignMap that unravels the influence of money on political discourse over time.

Acknowledgements

We thank the generous support of MIT IS&T, CSAIL, and the Department of Mathematics for their support of this series.