“The Science DMZ is a portion of the network, built at or near the campus or laboratory's local network perimeter that is designed such that the equipment, configuration, and security policies are optimized for high-performance scientific applications rather than for general-purpose business systems or ‘enterprise’ computing.” – ESnet, creators of the Science DMZ network architecture
USD's Science DMZ Network was funded through NSF award 1440681 “CC*IIE Networking Infrastructure: Science DMZ for Campus Bridging and Computationally Assisted Research”, a $498,813 grant spanning two years. The project delivered new network cyberinfrastructure for research data at USD.
High-throughput Big Data movement
USD’s Science DMZ project extends the capabilities of the Globus data movement and management system to all USD researchers. Globus provides an easy-to-use graphical interface for moving data into and out of USD's Vermillion campus. USD's Globus system resides on an high-performance data server called a Data Transfer Node, also funded through the Science DMZ project.
Big Data needs to be moved both quickly and securely. USD's Science DMZ project funded an advanced Intrusion Detection and Prevention system that monitors USD network traffic, protecting both the Science DMZ and the rest of the campus network.
- perfsonar1.usd.edu in USD’s edge network
- perfsonar2.usd.edu in USD's Science DMZ network
- perfsonar3.usd.edu in USD's datacenter core network
- perfsonar4.usd.edu in USD's campus core network
The Science DMZ supports scientific research across multiple disciplines at USD, including: particle physics, neuroscience, materials science, biomedical research, biology and renewable energy. Although diverse, these fields are bound by the common themes of cross-institutional collaboration, remote data access, and real time data analysis. USD scientists, for example, are investigating, the general properties of neutrinos and the nature of dark matter -- two of the most important pursuits in modern science. This data intensive effort, whose work involves state, federal and academic collaborators from around the world, requires a direct, high-speed path capable of hosting, sharing and analyzing large datasets to facilitate their work.
At the end of the project, a questionnaire was administered to faculty involved with this project. Responses show that data management needs and big data movement continue to grow and that access to onsite and offsite data is vital across disciplines (download the complete survey and results).
The following Before and After diagrams illustrate the new network architecture enabled by this project:
Before: 2014 Campus Network
After: Current Campus Network