In 2022, five Oklahoma institutions connected to the OneOklahoma Friction Free Network (OFFN), Oklahoma’s research network, through a National Science Foundation (NSF) Campus Cyberinfrastructure (CC*) grant awarded to the Oklahoma State Regents for Higher Education and OneNet. OFFN is a 10 and 100 Gbps research network that provides higher education institutions with a dedicated internet route that is much faster than traditional internet highways. This alternate pathway allows researchers to bypass slower traditional networks and transmit data at higher speeds. Administered by NSF’s Office of Advanced Cyberinfrastructure, the CC* grant program invests in campus-level cyberinfrastructure improvements for science applications and research projects. NSF has funded three awards for OSRHE and OneNet to expand the OFFN network.
“The OFFN connection provides opportunities for smaller institutions to access high-performance computing resources that previously were not a possibility, allowing for greater collaborations and scientific discovery,” said Brian Burkhart, OneNet’s senior director of network services. “The OFFN network runs parallel to OneNet’s network, and the extension of the OFFN network to the smaller institutions furthers OneNet’s mission of advancing research and education initiatives across the state.”
At Oklahoma Christian University (OC), one of the campuses connected to OFFN in 2022, joining the OFFN network has facilitated new connections and collaborations across the state for faculty and students to access supercomputing resources.
In spring 2022, students in OC’s big data management course put the new connection to work when they used the University of Oklahoma’s supercomputing cluster, Schooner, to perform a class assignment.
The big data management course is offered to graduate students as part of the Master in Computer Science program with a focus in data science. In this course, students generated half a gigabyte of data pertaining to randomized video games and transferred it onto the Schooner cluster using the OFFN network connection to do classification analysis. OC’s Associate Professor of Computer Science Dr. Robert Nix provided guidance and technical support for this collaboration.
“The OFFN network connection to our campus has provided us the ability to contain and transfer large amounts of data to the OU Schooner supercomputer, which then allows our students the ability to gain experience working with large data sets,” said David North, Chair of the OC computer science program.
The Master of Computer Science data science track includes three courses where students learn to use and practice a number of data science tools, such as writing a Hadoop MapReduce job to process files. This allows students to learn how to scale unstructured data across hundreds or thousands of commodity servers in an Apache Hadoop cluster.
Students also learn how to write an Apache Spark job to process the files in a different manner. Apache Spark is a data processing framework that can quickly perform processing tasks on very large data sets and can also distribute data processing tasks across multiple computers, either on its own or in tandem with other distributed computing tools.
Within the Apache Spark Machine Learning Library, students learn how to perform classification tasks on data. The library allows data students and scientists to focus on their data problems and models rather than solving the complexities surrounding distributed data, such as infrastructure and configurations.
While utilizing Schooner, OC students created their own accounts within the supercomputer to be able to access the data. Thanks to the guidance and training of the support teams at both OneNet and OU, students also learned how to successfully submit jobs to the cluster and retrieve the results.
After resounding success with this new course, OC plans to continue to leverage Schooner and OFFN resources in two additional data science projects courses in summer 2022. Students will generate larger data sets to work and transfer across OFFN.
OU’s assistance in accessing and setting up software on Schooner for OC students proved vital in teaching students about big data and how to best manage data in their careers. OneNet’s team provided critical support in setting up the data transfer via OFFN from OC’s campus to the OU Schooner supercomputer.
“This access has improved the experience our students gain in our graduate data science class. It allows them to experience working with real-world size data and access to the processing power of a supercomputer,” North explained.
North expressed OC’s appreciation for OneNet’s commitment to serving students in through learning initiatives and advancing this type of technology across Oklahoma.
“Having these resources available to students has expanded our thinking about what is possible in offering courses that pertain to career application,” North said.
OC is only one of many new OFFN connections. With the recent additions, OneNet will soon connect a total of 21 Oklahoma campuses to OFFN. OC’s collaboration with OU and the Schooner supercomputer demonstrates how important high-speed connectivity is for students to be able to access 21st century learning resources. This project is an excellent example of the innovative work taking place across the state of Oklahoma, thanks to the OFFN network.
University of Oklahoma Photo Credit: Jawanza Bassue