NS&D Monitor
5/9/2014
Lab Seeking Industry Collaborators for New Supercomputer
Lawrence Livermore National Laboratory is opening up a recently commissioned supercomputer for industry collaborators to test big data technologies, architectures and applications. A high performance computing cluster dubbed Catalyst—a collaboration between Livermore, Intel and Cray—came on line at Livermore last fall, and the lab’s High Performance Computing Innovation Center (HPCIC) is soliciting statements of interest from industry about potential work on Catalyst. “Our purpose is to use Catalyst as a test bed to develop optimization strategies for data-intensive computing,” HPCIC Director Fred Streitz said in a statement. “We believe that advancing big data technology is a key to accelerating the innovation that underpins our economic vitality and global competitiveness.”
When Catalyst was unveiled in the fall, project officials said it was designed to produce insights into the kind of technologies the Advanced Supercomputing Program will need over the next five-to-10 years to meet high performance simulation and big data computing mission needs. One significant feature of the machine is its increased storage capacity, which represents a major change from typical simulation-based computing architectures used at Department of Energy laboratories. Livermore said Catalyst’s expanded memory makes it perfect for addressing a variety of “big data” challenges, which it said could include bioinformatics, business analytics, machine learning, and natural language processing. “Over the next decade, global data volume is forecasted to reach more than 35 zettabytes (a trillion gigabytes),” Streitz said. “That enormous amount of unstructured data provides an opportunity. But how do we extract value and inform better decisions out of that wealth of raw information?”
Catalyst consists of two scalable units that are an upgrade on the Appro clusters that had been in use through the Tri-Lab Linux Capacity Cluster procurement several years ago. Catalyst is capable of 150 trillion floating operations per second (150 teraflop/s) and has 324 nodes, 7,776 cores and uses the latest 12-core Intel(R) Xeon(R) E5-2695v2 processors. It also includes 128 gigabytes of dynamic random access memory per node, 800 GB of non-volatile memory (NVRAM) per compute node, 3.2 terabytes of NVRAM per Lustre router node, and improved cluster networking with dual rail Quad Data Rate (QDR-80) Intel TrueScale fabrics.
