Researchers at Rice University are taking some BOLD steps to get a handle on the enormous amount of Big Data being generated by campus scientists.
That’s BOLD as in Big data and Optical Lightpaths-Driven networked system research infrastructure. While their acronym doesn’t quite work, the researchers are hoping the network itself will be the solution that the university needs to better hone massive amounts of information.
“Advances in computing and sensing technologies have led to a similar problem across many disciplines in science and engineering today,” said BOLD’s principal investigator T.S. Eugene Ng, an associate professor of computer science and of electrical and computer engineering at Rice. “Experiments produce mountains of data, and there is often no efficient way to process that data to make discoveries and solve problems.”
That’s a problem many institutions are facing as the amount of Big Data –the 2.5 quintillion bytes of information created every day through research, mobile devices, and online interactions – continues to grow.
Backed by a grant from the National Science Foundation, the researchers at Rice hope to create a customized, optical network to funnel the deluge of Big Data into the university’s super computers.
BOLD will accomplish this through the use of optical data-networking switches, which differ from typical electronic switches that are used at most data centers.
See Page 2 for details on BOLD’s complicated hybrid of networking switches and how powerful the system will be.
The optical switches use considerably less power and can facilitate large data rates, but they require extra effort to use as they must be first configured by moving microelectromechanical mirrors into place to establish the circuit.
BOLD will use a mix of both kinds of switches, Ng said, as well as a new kind of optical switch that works sans those moving mirrors. These switches will be created in a lab by Qianfan Xu, an assistant professor of electrical and computer engineering at Rice.
“To make use of these three types of technology, we need an intelligent layer that can analyze data flow and demand, all the way up to the application layer, and dynamically allocate network resources in the most efficient way,” Ng said.
With BOLD, Rice will be able to conduct operations and research that require the generation, loading, caching, recalling, and modification of tens to hundreds of terabytes of Big Data.
Just 10 terabytes of data is about the same size as the Library of Congress’ entire print collection.
Follow Jake New on Twitter at @eCN_Jake.