In 2013, China's Tianhe-2 became the fastest supercomputer in the world, achieving 33.8 petaflops, with a peak performance of 54.9 petaflops. It was nearly twice as powerful as its nearest competitor, the "Titan" machine at Oak Ridge National Laboratory in Tennessee, United States. Tianhe-2 showed great success in computational fluid dynamics (CFD), scramjet combustion and other aircraft simulations. Other research areas included genomics (population genetics) and biomedical applications. It continued to maintain the top spot for the next few years, with no other challengers, leading to concerns over a possible stagnation in the progress of supercomputing speeds. This was further compounded by trade restrictions – imposed by the United States government – which prevented China from using Intel's chip technology for upgrades. However, this actually boosted China's own processor development and production industry.
In 2016, Tianhe-2 received an extra 45 petaflops, thanks to a novel system architecture developed within China. The machine was renamed Tianhe-2A and became the first supercomputer to demonstrate a peak performance of 100 petaflops, or 100 quadrillion floating point operations per second.* Alongside this upgraded version was a second machine, known as Shenwei-x. This too was capable of 100 petaflops. Both machines would reign at the top of the list for only a short while, however, as the United States was regaining its lead in supercomputer research. A project called "Aurora" was being developed at the Argonne Leadership Computing Facility (180 petaflops), along with IBM's "Summit" (150 petaflops), the latter having an optional upgrade to 300 petaflops. This would be followed by the first exascale computers (1,000 petaflops) towards the end of the decade.
source
