Cray Laps Itself in Supercomputing Race
Cray continues to push the boundaries of supercomputing. We previously wrote about the delivery of the Titan to Oak Ridge National Laboratory, and that system just nabbed the No. 1 spot on the Top500 list of the world’s fastest supercomputers. Titan (a Cray XK7) reached 17.59 petaflops on the Linpack benchmark, but Cray is already shipping systems that promise to leave Titan in the dust. The company has launched its next-generation XC30 supercomputer (previously code-named Cascade) that is designed to achieve HPC workloads of more than 100 petaflops and scale up to 1 million cores.
A number of major customers have already signed contracts to purchase XC30s (to the tune of around $100 million), including the Swiss National Supercomputing Center, the Finnish IT Center for Science, the Academic Center for Computing and Media Studies at Kyoto University in Japan, and the Department of Energy’s National Energy Research Scientific Computing Center (NERSC) in Berkeley, CA. Early shipments have already started, and the systems are expected to be generally available in Q1 2013.
What’s Cray doing with all that money? Expanding. The company is acquiring HPC cluster systems specialist Appro International for $25 million in cash.