How Fast Database MapD Powers Real-Time Twitter Maps

by   |   April 26, 2013 12:23 pm   |   0 Comments

CAMBRIDGE, Mass.—Todd Mostak had a turbulent experience with trying to process and plot tweets during the Arab Spring for his master’s thesis for the Center for Middle Eastern Studies at Harvard. So while taking a course on databases at MIT, he created a massively parallel database using inexpensive graphical processing units (GPUs) that could attack data problems with brute force processing power.

You can read more about Mostak’s development journey on Data Informed.

In this video Mostak explains his database, called MapD, and shows off his visualization tool TweetMap, an early use of the GPU-based system. TweetMap is in an alpha stage of development  and is hosted by Harvard’s open source WorldMap GIS project. The visualization system can instantly visualize dynamic heat maps on geolocated tweets, and currently is available to the public to explore  125 million geolocated tweets sent in December 2012.

Mostak said the TweetMap was a bit of a “low hanging fruit” for MapD; because of it’s GPU makeup, crunching and graphically rendering map data is a logical application. But Mostak said there are many possible use cases for the system, from processing and trend detection  on data stored in graph databases to machine learning.

Mostak recently created his own demo video, showing the reaction on Twitter on of the April 15 bombings at the 2013 Boston Marathon.

Email Staff Writer Ian B. Murphy at

Also on Data Informed:

Hack/Reduce Boston to foster big data talent.

 SAS CEO Jim Goodnight on modernizing analytics for massively parallel systems.

SAP pitches HANA platform to analytics startups.

Tags: , , , , ,

Post a Comment

Your email is never published nor shared. Required fields are marked *

You may use these HTML tags and attributes: <a href="" title=""> <abbr title=""> <acronym title=""> <b> <blockquote cite=""> <cite> <code> <del datetime=""> <em> <i> <q cite=""> <s> <strike> <strong>