Intel announced today 6th in a series of Intel Science and Technology Centers (ISTCs) with the opening of the ISTC for Big Data. The center will be headquartered at MIT, with a community of collaborating institutions including Brown University, Portland State University, University of Washington, University of Tennessee at Knoxville, University of California at Santa Barbara, and Stanford University
We opened this center to address the unprecedented growth in digital data and the issues that it entails. In the future, this is likely to accelerate even further through more pervasive use of digital media, ubiquitous connectivity, growing use of smart mobile devices, growing penetration of geo-spatial mapping and round-the-clock sensing. This data flood will be large in size, multi-dimensional and highly correlated. Storing and providing unique visualization techniques to efficiently parse and extract meaningful information from this data is key, and it is the ultimate goal of the center.
With this in mind, this new ISTC for Big Data will explore data analytics to support data-intensive discovery including database management, analytics and visualization support for massive datasets and their implications for computer architectures.
Expanding on this, the research at this center will focus on exploring the challenges and opportunities associated with Big Data including
A) Designing and prototyping hardware and software for storing, managing, processing, understanding, and visualizing “big data”. This includes user interfaces and data visualization.
B) Discovering novel algorithms and scalable, co-designed architectural alternatives for effectively dealing with the three inherent attributes of big data: massive, unstructured, and dynamic;
C) Innovative ways of exploiting modern processor technology trends such as, multicore, manycore, and emerging non-volatile memory technologies
The research could have impact in several areas. Financial services applications are an example of business with big data challenges, as they look for patterns to try to make sense of the huge streams of data.
In science, there’s already a huge volume of data that needs to be processed in areas like astronomy (to make sense of Petabytes of data from telescopes) or genetics (with experiments on gene expression with the hope of creating more customized treatments for people), as well as medical collaboration and diagnosis.
Another area is “urban-scale sensing” where cities have the potential to generate massive amounts of data from GPS-enabled vehicles and other devices, electricity and water monitors, and weather and air quality systems. The research would help with storing the data and making sense of it so that the city managers and inhabitants can benefit.
We expect synergy between this new center and the other ISTC centers, such as the ISTC for Cloud Computing and ISTC for Embedded Computing.
The leaders of the center are Dr. Sam Madden (Academic PI) and Dr. Michael Stonebreaker (Academic Co-PI) both at MIT, Dr. Pradeep Dubey (Intel PI), and Dr. Jeff Parkhurst (Intel program director).