Carnegie Mellon Builds New Computing Cluster for Education and Research

Carnegie Mellon University (CMU) is building a new computing cluster, named Narwhal, which will be used for education and research related to large-scale computer systems.

The Narwhal cluster will include 448 blade computers, 1,792 processor cores and more than 400 magnetic disks. Narwhal's blade computers were salvaged from Cerrillos, a former Los Alamos National Laboratory (LANL) supercomputer that was decommissioned last year. Cerrillos contained a whopping 14,400 processor cores, but even Narwhal's fraction of Cerrillos's core capabilities "is more capable than most educational computing resources," said Garth Gibson, professor of computer science and principal investigator for Narwhal, in a prepared statement.

Cerrillos was related to the Roadrunner supercomputer — the first computer to break the petaflop barrier by performing more than one million billion calculations per second. Cerrillos and Roadrunner shared a hybrid architecture that used both AMD Opteron processors and special graphics processors from IBM, called Cells, to increase overall processing speed. However, Narwhal does not use Cerrillos's Cell processors "because the academic parallel and distributed computing goals for Narwhal were better met with more blades and fewer Cell co-processors," according to the CMU.

Researchers and students at CMU will use the Narwhal cluster "to experiment with parallel computing applications and infrastructure, controlling and instrumenting all software down to the bare metal, at a scale that's an order of magnitude larger than most university clusters," according to the university.

"With Narwhal, we open a new front — assistance with large-scale computer systems software education," said Gibson in a prepared statement.

The Narwhal cluster is named after the Arctic whale known for the long tusk protruding from its head. The name was chosen in honor of the Parallel Reconfigurable Observational Environment's (PRObE) tradition of selecting names with Alaskan themes. Garth Gibson, principal investigator for Narwhal, was part of the team that established PRObE.

CMU's Parallel Data Lab will set up the Narwhal cluster this summer in the Robert Mehrabian Collaborative Innovation Center (CIC) on the CMU campus.

About the Author

Leila Meyer is a technology writer based in British Columbia. She can be reached at [email protected].

Featured

  •  black graduation cap with a glowing blue AI brain circuit symbol on top

    Report: AI Is a Must for Modern Learners

    A new report from VitalSource identifies a growing demand among learners for AI tools, declaring that "AI isn't just a nice-to-have; it's a must."

  • AI-powered individual working calmly on one side and a burnt-out person slumped over a laptop on the other

    Researchers: AI's Productivity Gains Come at a Cost

    A recent academic study found that as organizations adopt AI tools, they're not just streamlining workflows — they're piling on new demands. Researchers suggested that "AI technostress" is driving burnout and disrupting personal lives, even as organizations hail productivity gains.

  • illustration of a football stadium with helmet on the left and laptop with ed tech icons on the right

    The 2025 NFL Draft and Ed Tech Selection: A Strategic Parallel

    In the fast-evolving landscape of collegiate football, the NFL, and higher education, one might not immediately draw connections between the 2025 NFL Draft and the selection of proper educational technology for a college campus. However, upon closer examination, both processes share striking similarities: a rigorous assessment of needs, long-term strategic impact, talent or tool evaluation, financial considerations, and adaptability to a dynamic future.

  • young woman using a smartphone, with digital AI and chat icons overlaid in a blurred academic setting

    Duolingo Embraces AI in Push for Scalable Learning

    Learning platform Duolingo has officially declared itself "AI-first," aiming to make learning replicable, scalable, and always available.