September 06, 2023
By Brittany Hilderbrand | Division of IT
A MizzouForward initiative is opening doors for students, including Ethan Koen, an MU senior majoring in information technology, to get real-world experience working on one of the largest supercomputers in the Midwest.
While working with the Division of IT Research Support Solutions team, Koen had the opportunity to assist in bringing 100% power to Hellbender – the university’s new HPC environment. This project is part of the MizzouForward initiative and an associated $7 million investment to develop next-generation computing resources and strengthen the university’s research data ecosystem.
For many researchers, the ability to compute and store their data determines how quickly they can produce their research and often requires collaboration with the IT RSS team.
In April 2023, because of delays in equipment delivery necessary for full power, technologists were able to bring Hellbender online at 50% capacity but as researchers were onboarded and larger jobs were run, the need for the full capacity of the system became more urgent.
The new reality technologists faced was that the power limitations were hampering researcher productivity sooner than anticipated. A plan was launched to find a temporary workaround and enable Hellbender’s full capacity before the rest of the permanent power infrastructure was in place.
Eli Riekeberg, system administrator with RSS, explained the job required the team to create artificial loads that would push every program to 100%, label all the cables so the power loads are manageable and then balance the number of loads placed on the individual power distribution units.
Throughout the implementation stages of this project Koen had an opportunity to put his academic knowledge to the test.
“As a senior, I have already taken most of the IT classes at MU so it has been very good for me to have hands-on experience through my time with ITRSS,” Koen said. “One of my biggest takeaways from working on this project is how much planning it takes to put together a cluster like Hellbender. Another takeaway is how dynamic these plans must be; things can change quickly so we have to adapt.”
Throughout the phases of powering on Hellbender, the team experienced various setbacks including shipping delays and backorders for adequate power distribution units (industry power cords). The technologists drive to push pass these hurdles and utilize the resources they had while working side-by side with various members of MU & S&T IT teams.
Koen’s ability to adapt and learn from RSS team members has allowed him to work on other high-level projects where he has gained valuable experience to add to his resume.
“Ethan has been involved in maintaining Hellbender and the legacy HPC cluster known as Lewis, including coordinating with major vendors like Dell,” said Matt Stanley, system administrator in RSS. “In many ways he’s gone from being a student to being as valuable of a member on the team as any of us.”
“Having the opportunity to configure the power cables for one of the largest supercomputers in the Midwest is a very rare experience,” added John Harrison, RSS X-team Manager. “Experience like this doesn’t come around all the time. This is an impactful real-world application.”
Be part of a collaborative engineering program that offers one-of-a-kind opportunities. Apply today!