Among them are the two fastest computers in the world (both in the US), and the fastest in Europe and Japan.
The two leading supercomputers are Oak Ridge National Laboratory's Summit (200,794.9 Tflops) and Lawrence Livermore National Laboratory's Sierra (125,712.0 Tflops), both of which use IBM Power CPUs as well as Nvidia Volta GPUs.
Europe's fastest is the Swiss National Supercomputing Centre's Piz Daint (100,678.7 Tflops, Intel Xeon plus Nvidia Tesla), which came in at number five.
Two places behind that is Japan's fastest supercomputer, the National Institute of Advanced Industrial Science and Technology's ABCI (32,576.6 Tflops, Intel Xeon and Nvidia Tesla).
One of the newcomers to the Top500 is the Nvidia DGX-2 Pod, a cluster of 36 DGX-2 systems (pictured) delivering more than three double-precision petaflops. Nvidia estimates that a cluster of just 11 DGX-2s would qualify for the latest Top500 list.
Nvidia GPUs are also used in eight out of the top ten and 22 of the top 25 supercomputers on the Green500 list. Nvidia's own DGX SaturnV Volta system took second place by achieving 15.113 Gflops/watt, behind Shoubu system B, a Pezy/Exascaler system at Japan's Advanced Center for Computing and Communication which recorded 17.604Gflops/watt.
"This is a breakout year for Nvidia in the world of supercomputing," said Nvidia founder and chief executive Jensen Huang.
"With the end of Moore's law, a new HPC market has emerged, fuelled by new AI and machine learning workloads. These rely as never before on our high performance, highly efficient GPU platform to provide the power required to address the most challenging problems in science and society."