While the US Lawrence Livermore National Laboratory is in the forefront of developing exascale technology for the next generation of supercomputers through the Coral procurement initiative, its parent organisation, the National Nuclear Security Administration (NNSA), has just announced a $39 million investment in commodity computing clusters to conduct nuclear weapons calculations at all three of the NNSA’s laboratories.
The contract was awarded by Lawrence Livermore as part of NNSA’s tri-lab Advanced Simulation and Computing (ASC) programme to Penguin Computing. Penguin will provide more than 7 Petaflops (quadrillion floating operations per second) of computing capability to Los Alamos, Sandia, and Lawrence Livermore. These commodity technology systems (CTS) are designed to run a large number of jobs simultaneously on a single system and they are intended to take some of the pressure off the laboratories’ more powerful supercomputers by running the more routine calculations that the labs need to carry out.
Doug Wade, head of NNSA’s ASC programme, said: ‘These computing clusters will provide computing capacity for NNSA’s day-to-day work at the three labs managing the nation’s nuclear deterrent. This tri-lab effort will help reduce costs, increase operational efficiencies, and facilitate collaborations.
The decision to buy commodity clusters for this aspect of the nuclear weapons work will allow the NNSA’s more powerful ‘advanced technology system’ (ATS) supercomputers to concentrate on the largest and most complex calculations. High performance computing is a cornerstone of NNSA’s ‘Stockpile Stewardship programme’ to ensure the safety, security, and reliability of the ageing US nuclear deterrent without actual physical testing. The ASC programme combines the computing expertise and resources at the three labs to advance the computing capabilities.
The new capacity computing systems, called the Commodity Technology Systems – 1 (CTS-1), will be the third joint procurement of this type and will replace those procured in 2011, which are now nearing retirement. This tri-lab procurement model reduces costs through economies of scale based on standardised hardware and software environments at the three labs.
Under terms of the contract, computing clusters built of ‘scalable units’ (SUs) will be delivered to each of the laboratories between April 2016 and September 2018. Each scalable unit represents approximately 200 Teraflops of computing power. These SUs are designed to be connected to create more powerful systems. Scalable units will be divided among the three labs with each configuring the SUs into clusters according to their own workflows.
CTS-1 clusters will support NNSA’s Life Extension Programme and investigations into technical issues related to aging weapons systems, efforts critical to ensuring the safety security and reliability of the nuclear weapons in the stockpile as they age well beyond their intended deployment life.
