Open-Compute supercomputers enlisted for nuclear security
Connecting state and local government leaders
The National Nuclear Security Administration has contracted for supercomputing clusters based on the Open-Compute Project, the Facebook-led open source hardware and data center design initiative.
The National Nuclear Security Administration will be bolstering its high-performance computing capacity at its three national security laboratories with Tundra Extreme Scale series systems from Penguin Computing, a leading developer of high-performance Linux cluster computing systems. These supercomputing clusters will be one of the world's largest installations of hardware based on the Open-Compute Project, the Facebook-led open source hardware and data center design initiative.
Under the contract for NNSA’s tri-lab Advanced Simulation and Computing (ASC) program, Penguin Computing will deliver over seven petaflops of capacity to the Los Alamos, Sandia and Lawrence Livermore national laboratories. That processing power will be used to support NNSA’s efforts to ensure the safety security and reliability of the nuclear weapons in the stockpile as they age well beyond their intended deployment life.
“These computing clusters will provide needed computing capacity for NNSA’s day-to-day work at the three labs managing the nation’s nuclear deterrent,” said Doug Wade, head of NNSA’s ASC program. They are designed to run a large number of computing jobs simultaneously on a single system, which allows NNSA’s more powerful supercomputers to be dedicated to the largest and most complex calculations critical to stockpile stewardship.
Under terms of the contract, computing clusters built of “scalable units” will be delivered to each of the laboratories between April 2016 and September 2018. Each SU represents approximately 200 teraflops of computing power, and they are designed to be connected, much like Legos, to create more powerful systems. The SUs will be divided among the three labs with each configuring the SUs into clusters according to its mission needs.
The Tundra system will replace those machines procured in 2011, which are now nearing retirement. This is the agency’s third joint procurement, a model that reduces costs through economies of scale based on standardized hardware and software environments at the three labs.
Penguin Computing will begin delivery of this Tundra Extreme Scale solution in Q1 of 2016, with completion of the installation over a three-year period.