Skip to main content

Bland delivered world-leading supercomputing systems in his tenure at DOE’s Oak Ridge Leadership Computing Facility

“No risk, no reward.” It’s a familiar sentiment in business and entrepreneurial ventures. Risk drives innovation and propels organizations to new heights. It’s certainly familiar to Arthur “Buddy” Bland, former project director at the Oak Ridge Leadership Computing Facility (OLCF) who retired this month after 40 years of faithful service to the US Department of Energy (DOE) and the nation delivering some of the most powerful supercomputers in the world.

Bland delivered and deployed the Jaguar, Titan, and Summit supercomputers—mammoth machines that earned number one spots on the TOP500 list of the fastest supercomputers in the world at different times. All three were housed at the OLCF, a DOE Office of Science User Facility at DOE’s Oak Ridge National Laboratory (ORNL). For Bland, risk was especially key in managing the deployment of the Cray XK7 Titan, the first large-scale GPU-CPU system to break the 10-petaflop performance barrier.

“Choosing a GPU-accelerated system was considered a risky choice,” Bland said in 2019 amid Titan’s decommissioning. “A DOE independent project review committee insisted that we demonstrate that our users would be able to effectively use Titan for the broad range of modeling and simulation applications we support. We spent 6 months working with Cray, NVIDIA, and our users to convince the reviewers, DOE, and ourselves that GPUs would deliver what we needed.”

Now, after 36 years at ORNL—14 of which were spent delivering the OLCF’s top systems—Bland is retiring from his computing career, leaving behind a legacy of relationship building, collaboration, and perseverance. He was recently presented the DOE Secretary’s Exceptional Service Award—the highest of the Secretary’s Departure Awards—to recognize his exceptional service to DOE and the nation delivering extraordinary leadership in high-performance computing (HPC). Former DOE Secretary of Energy Dan Brouillette presented the award to Bland virtually in December.

“It is important to understand that when we say that Oak Ridge National Laboratory led the United States into a new era of high-performance scientific computing, it is because of supercomputers that were prepared, delivered, and put into operation at the direction of Buddy Bland,” said Thomas Zacharia, who led the creation of the OLCF and is now the laboratory director of ORNL. “Buddy worked tirelessly, identified and trained talented team members, solved problems that would have overwhelmed less determined individuals, and brought a series of world-leading systems online. It is impossible to talk about ORNL today without explaining its role in leadership computing, and that is due in large part to the contributions of Buddy Bland,” Zacharia said.

Buddy Bland, OLCF Project Director, July 3, 2019. Image Credit: Carlos Jones

Rising to the TOP500

When Bland arrived at ORNL in 1984 as a system programmer and administrator for ORNL’s Cray X-MP system, he couldn’t have imagined that he would one day direct supercomputing projects as massive as the OLCF’s Summit project. Since Bland arrived at ORNL, the OLCF has seen more than a billionfold increase in computing power.

After managing the Cray X-MP and a series of other systems—three iPSC (Intel Personal SuperComputer) systems and the first multiprocessor system from Kendall Square Research, the KSR-1—Bland became the director of operations at the Center for Computational Sciences (CCS), which Congress awarded to ORNL in 1992. CCS would establish leadership-class computing facilities within DOE as part of the High-Performance Computing Act of 1991. But Bland’s biggest contributions came after the OLCF was formed in 2004. The OLCF’s goal was to provide researchers from government, academia, and industry with access to HPC resources, with individual programs receiving 100 times more computing capability than was available at other facilities at the time.

Nobel Laureate Albert Fert receives a tour of the computer room with Buddy Bland and Jeff Nichols, November 4, 2013. Image Credit: Jason Richards, ORNL

In 2006, Bland became the project director of the OLCF. His team’s first goal was to develop and deploy the first open-access petaflops computer—a system that would be 100 times more powerful than the leading systems at the time. The team’s efforts proved successful when the Cray XT5 Jaguar supercomputer was deployed in 2008. It became the first system to run a scientific application at a sustained petaflop that same year. After upgrades to the machine, it rose to number one on the TOP500 list in 2009.

Barbara Helland, associate director of the DOE Office of Science Advanced Scientific Computing Research program, remembers well the first projects Bland delivered at the OLCF.

“It was clear from the beginning of the Leadership Computing Facilities that Buddy had the experience, knowledge, and calm presence to deliver number one machines,” Helland said. “This year, the ORNL team will deliver the first exascale system for the nation, a system that will benefit from the foundation that he has laid.”

Bland and his team went on to deliver the 27-petaflop Titan system in 2012 and the OLCF’s current 200-petaflop Summit system in 2018, both former number one systems on the TOP500 list.

“If you can point to any one person who has been there for the duration of the OLCF’s history, who has been the leader when things were at their lowest and when things were at their best, it’s been Buddy,” said Don Maxwell, who leads the HPC Scalable Systems Group in the National Center for Computational Sciences. Maxwell has worked with Bland for nearly his entire career. “He is truly one of the biggest reasons that we are so successful.”

A legacy of relationships

Although Bland has retired, the lessons he imparted to his team throughout the years remain. His management approaches focused on building relationships, engaging in collaboration, and finding solutions to the most imminent problems first.

Buddy Bland and Jim Hack with the Summit supercomputer, April 11, 2018. Image Credit: Jason Richards, ORNL

“He used to call it ‘management by walking around,’” said ORNL’s Kathlyn Boudwin, who worked as deputy project director under Bland. “He would just sit in people’s offices, and you always had this feeling that you could tell Buddy anything. Not only was he very technically competent, he was also just also a very warm person.”

Bland has also been mentoring the current project director for the Frontier project, Justin Whitt, throughout the last 5 years. Frontier is expected to be deployed this year as a 1.5-exaflop system at the OLCF.

“I have learned many things from Buddy about building supercomputers in the past few years,” Whitt said. “Buddy’s work has enabled countless researchers to accomplish their goals and produce impactful science at unprecedented scales, and the organization that he helped build will continue to do so for years to come.”

Although Whitt appreciates Bland’s technical expertise and experience, he believes Bland’s true legacy lies in the relationships he built.

“I’ve learned that it takes a broad set of relationships, an intense focus, and hard work to be successful,” Whitt said. “Buddy’s legacy is in what he established at the OLCF: communication and coordination among the highly talented technical teams at the center, broad support from across ORNL, a close partnership and trust with DOE, and long-term collaborations and shared successes with HPC companies.”

UT-Battelle LLC manages Oak Ridge National Laboratory for DOE’s Office of Science, the single largest supporter of basic research in the physical sciences in the United States. DOE’s Office of Science is working to address some of the most pressing challenges of our time. For more information, visit https://energy.gov/science.

Rachel McDowell

Rachel McDowell is a science writer for the Oak Ridge Leadership Computing Facility.