20 Jun 2017 Tokyo - On June 19, the K computer took the first place for the second consecutive time in the HPCG benchmark, a new index developed to create a more realistic view of supercomputer performance compared to the commonly used LINPACK benchmark. This success, which surpasses the second place achieved in 2014 and 2015, was made possible by subsequent improvements of the performance of the system and applications.
The HPCG (High Performance Conjugate Gradient) benchmark measures how fast a computer can solve symmetric sparse linear system equations using the conjugate gradient method preconditioned with a multi-grid symmetric Gauss-Seidel smoother. Problems of this type are typically encountered in actual engineering and industrial applications, and require a balance between calculation performance, memory performance and communication performance, unlike LINPACK, which looks at calculation speed alone.
For this result, all of the K computer's 82,944 compute nodes were used, achieving a performance of 602 teraflops. This figure is higher than the supercomputers that placed higher than the K computer in the TOP500 rankings, demonstrating outstanding performance in various science and engineering fields.
According to Mike Heroux of Sandia National Laboratories, who developed the HPCG benchmark: "The HPCG benchmark is very demanding. A good score requires strong and versatile memory system performance, excellent interconnect network performance at scale, and an overall balanced system. The top ranking of the K computer is no surprise."
The award will be presented at ISC High Performance 2017 in Frankfurt, Germany.