OLCF, NCCS staff part of team delivering course curriculum
Though invisible to the average person, data centers have become ubiquitous in the information age, housing the computing systems that enable much of the information technology we’ve come to rely on in our day-to-day lives. It’s estimated there are about 3 million data centers in the United States alone—some small enough to stuff in a closet, others occupying entire buildings.
Fueled by the Internet economy and the need for more complex computing services and environments, data centers are continuing to grow and evolve at a fast pace—so fast, in fact, that universities are scrambling to equip students with in-demand data center skills.
This fall, the US Department of Energy’s (DOE’s) Oak Ridge National Laboratory (ORNL) partnered with the University of Tennessee (UT) to offer a minor-degree program in data center technology and management, one of the first offerings of its kind in the country. The program is anchored by a new course, ECE 463: Introduction to Datacenters, designed to give students a broad overview of designing and operating such a center.
ORNL staff members developed the senior-level course in collaboration with UT College of Engineering professor Mark Dean after an ORNL strategic partner identified a need for employees who could bridge both the facilities and operational aspects of running a data center. In addition to developing the course curriculum, ORNL staff members are also serving as guest lecturers.
“There’s an emerging gap in skilled managers with sufficient understanding of both computing and facilities needed to tackle the unique challenges of large data center environments, and this course is the beginning of an effort to bridge that gap,” said Jim Serafin, complex manager of Computational Facilities at ORNL. “It’s a good way for the lab to reach out to the best and brightest at UT and expose them to these unique opportunities. It’s also a great experience for our folks to share all the knowledge we’ve acquired operating and managing high-performance computing systems and their power and cooling infrastructure.”
Throughout the 15-week semester, 11 ORNL staff members cover data center topics based on their area of expertise. Topics include data center requirements, design, management, and methodologies, including reliability, security, network systems, storage systems, industrial design, systems management, application management, operations, logistics, and energy efficiency. The diversity of speakers from both the facilities side, which is responsible for the power, cooling, and industrial control systems that maintain data centers, and the computing side, which is focused on system performance, gives students broad insight into data center operations.
At the end of the course, students apply what they’ve learned by designing their own data center based on a mock request for proposal. The assignment requires students to develop a design strategy, create a floor plan, and account for the power, cooling, and networking needs of the customer.
For Stephen McNally, operations manager at the Oak Ridge Leadership Computing Facility (OLCF), a DOE Office of Science User Facility at ORNL, his experience overseeing the day-to-day maintenance of the OLCF’s compute and storage facilities provided the real-world backdrop that illustrated key concepts to students.
“During my lectures, I got the chance to talk about real projects being carried out at the OLCF and our role enabling leadership-class science,” McNally said. “All of us that are presenting aren’t just telling students what to do; we’re telling them about our experience, which students really seem to respond to.”
Kathlyn Boudwin, associate projects director for the Computing and Computational Sciences Directorate, echoed McNally’s sentiment, noting the growing role of project management in managing high-performance computing resources.
“The OLCF’s experience with project management in acquiring and standing up supercomputers gave real examples to students that they could wrap their minds around,” Boudwin said. “Project management is the glue that holds the engineering and technical components of what they’re learning together.”
In addition to Serafin, McNally, and Boudwin, the ORNL team contributing to the course includes David Grant, Rick Griffin, Bart Hammontree, Saeed Ghezawi, Mallikarjun Shankar, Jim Rogers, Craig Thomas, and Scott Milliken.
Oak Ridge National Laboratory is supported by the US Department of Energy’s Office of Science. The single largest supporter of basic research in the physical sciences in the United States, the Office of Science is working to address some of the most pressing challenges of our time. For more information, please visit science.energy.gov.