A supercomputing cluster in China has laid claim to the title of fastest in the world. The recently-upgraded Tianhe-1 cluster has reported a top performance of 2.507 petaflops, more than 2.5 trillion calculations per second. The benchmark blows past a record set by the Oak Ridge National Laboratories Cray Jaguar cluster, which topped out at a speed of 1.5 petaflops. The announcement of the 2.5 petaflop mark comes just two years after IBM’s RoadRunner system clipped the 1 petaflop barrier for the first time. Located at China’s National Supercomputer Center in Tianjin, the system was designed by the country’s National University of Defense Technology. Prior to the update, the Tianhe-1 cluster had ranked seventh on the Top500 supercomputing list. The system uses a combination of 14,336 Intel Xeon processors and 7.168 Nvidia Tesla graphics processors. Nvidia said that by integrating graphics chips into the cluster, a technique known as general processing over GPU (GPGPU) the system is able to cut its energy consumption from 14 megawatts to just over 4 megawatts.