CC* Networking Infrastructure: Improving Network Infrastructure to Enable Large Scale Scientific Data Flows and Collaboration

William Marsh Rice University, Houston TX

Investigators

Abstract

Campus networks are required to protect information and inspect data flows to safeguard security and privacy. However, researchers need open and unfettered access to large data flows and instruments across the globe to reduce time to discovery. The Rice University network is a shared resource that not only needs to support the administrative and teaching functions but also enable scientists to use that network in new and innovative ways for research. Five key data-intensive application teams act as drivers of the new extension of network functionality and are providing feedback to the technical design staff. These application areas include earth and atmospheric sciences; urban data science; computational biosciences and neuroengineering; particle physics and distributed cluster computing. These applications build on long-term science investments aimed, amongst others, at understanding seismic events, the weather patterns in the Gulf regions and beyond, as well as urban trends in large, diverse cities such as Houston, TX. The basic model adopted by the project is "the science DMZ." A Science DMZ is "a portion of the network, built at or near the campus local network perimeter designed so that equipment, configuration, and security policies are optimized for high-performance scientific applications rather than for general-purpose business systems." This approach allows Rice to aggressively upgrade its network capacity for greatly enhanced science data access. This project supports 100 GB/s flows between the data transfer facilities at our off-campus data center and national and international R&E data repositories and takes advantage of SDN (Openflow) mechanisms.