Rapid advancements in data-intensive scientific instrumentation have greatly outpaced the capability of campus networking infrastructures to effectively connect big data-producing facilities to powerful computing and storage systems. This delays the analysis, use and dissemination of a huge amount of valuable scientific data for discovery and innovation in multiple science and engineering domains. Purdue's project bridges this gap by improving the campus high-speed network infrastructure between five big data instrument facilities and the centrally supported cyberinfrastructure (CI) consisting of supercomputers, large storage systems and network connections to outside the campus. The facilities support accelerated research in new materials, understanding brain functions and viruses, monitoring lower atmospheric weather, employing geospatial data in teaching and public engagement, and secure computing. The project builds upon a strong partnership between campus information technology experts and domain scientists. It also helps prepare the next generation of CI professionals by pairing student workers with the campus CI experts.
The project adds high-speed Science DMZ connections to the five selected big data facilities enabling high-volume, high-velocity, or interactive science data flows to both the campus research cyberinfrastructure and off campus facilities. Existing Cisco Nexus 7710 network switches at the distribution layer are augmented with new 40-Gigabit networking modules, increasing bandwidth to 20-80 Gigabits per second. This should yield a significant increase in research productivity, and more timely publication and dissemination of research data and results, through faster data transfer, easier access to the central computing infrastructure, and more effectively utilizing Purdue's 100-Gigabit WAN connections and research network peerings.
This award reflects NSF's statutory mission and has been deemed worthy of support through evaluation using the Foundation's intellectual merit and broader impacts review criteria.
