Rice University researchers take "BOLD" approach to big data
- By Rutrell Yasin
- Sep 24, 2013
Editor's note: This story was updated to correct the amount of the NSF funding.
Computer networking researchers at Rice University have a bold idea for how to handle the massive amounts of data generated by scientists around campus. Armed with a grant from the National Science Foundation, the team will create a customized, energy-efficient optical network that can feed researchers’ data to Rice's supercomputers.
The project could also show how future nationwide networking infrastructures can use optical networks as shared cloud resources, according to the NSF, which awarded the university a three-year, $900,000 grant to build the Big Data and Optical Lightpaths-Driven Networked Systems Research Infrastructure or BOLD.
Advances in computing and sensing technologies have led to a big data problem across many disciplines in science and engineering today, said T.S. Eugene Ng, associate professor of computer science and of electrical and computer engineering at Rice.
Simply put, there is a dearth of supporting technologies capable of handling the outpouring of data generated by today’s big science and research programs. Optical networks might offer some of the answers.
"Experiments produce mountains of data, and there is often no efficient way to process that data to make discoveries and solve problems,” Ng said. BOLD will help researches solve this big data problem, said Ng, who is BOLD’s principal investigator.
The challenges go beyond just moving data, according to Ng. "We also need to develop transformative ideas in the network control software, operating systems and applications so that they can keep up with a faster network.Above all, for this network design to be appealing to industry, it has to be energy-efficient, scalable and nonintrusive to the end user,” he explained.
BOLD will take advantage of optical data-networking switches, which have much higher capacity than the electronic switches typically used in Internet data centers.
Optical networking devices also consume very little power and can support enormous data rates. But they must first be configured by moving micro-electromechanical mirrors into position to establish a circuit, Ng said. Electronic switches do not have moving parts and so do not experience mirror-positioning delays.
BOLD will merge the best of both worlds, combining both electronic and optical switches. The hybrid network will also consist of silicon-photonic switches without the moving parts and delays of traditional switches, Ng said.
“Results from the inter-disciplinary research enabled by BOLD will lead to future big data processing system architectures that dramatically speed up a wide range of computational scientific discoveries,” according to the NSF. Additionally, BOLD will serve as a platform for the training and education of many undergraduate and graduate students in leading edge big data-driven research.