DARPA Funds Python Big Data Effort

The Defense Advanced Research Projects Agency, which is spending $100 million over four years to advance big data technologies, recently awarded $3 million to develop data analytics and data processing libraries for popular computer programming language Python.

The funding, awarded to data visualization and analytics company Continuum Analytics, will go toward the development of a scientific computing library for Python called Blaze and a visualization system called Bokeh, Continuum announced in a blog post.

The funding, which was awarded in December, comes as part of XDATA, a $100 million DARPA research and development project that aims to "develop computational techniques and software tools for processing and analyzing the vast amount of mission-oriented information for Defense activities." XDATA, in turn, is but one piece of larger big data efforts by the Department of Defense and the federal government.

[ What else is DARPA up to? Read DARPA: Your Tech Will Self-Destruct. ]

Blaze will extend the current NumPy mathematical computing library and the SciPy scientific computing library for Python and will make those libraries more useful with big data. Key developers of those technologies, including NumPy's original author Travis Oliphant, will be involved with Blaze.

According to Continuum, Blaze will be "designed to handle out-of-core computations on large data sets that exceed the system memory capacity, as well as on distributed and streaming data."

Bokeh, meanwhile, is a Python library for big data visualization, what Continuum terms a "scalable, interactive and easy-to-use visualization system" for big data sets. Bokeh will integrate numerous visualization approaches, according to Continuum, including the Stencil visualization model and Grammar of Graphics.

A number of other big data efforts have also been recipients of DARPA's largesse.

In November, data visualization and processing software company Kitware announced a $4 million award to work with software company KnowledgeVis and universities on an open source data aggregation, querying and visualization toolkit called the Visualization Design Environment.

Then, in December, Georgia Institute of Technology announced that it had received a $2.7 million award to work on scalable machine-learning technologies and distributed computing architectures to rapidly process data analytics algorithms, and Scientific Systems Company, Inc., won an undisclosed amount of funding for new machine learning software.

Last month, database developer and software consultancy SYSTAP won $2 million as part of XDATA to build an open source graph analytics platform for use with GPU compute clusters.