The 17th Graph500 list — which ranks supercomputers based on how quickly they can build knowledge from massive-scale data sets — was released Nov. 13 at Supercomputing 2018 (SC’18), with RIKEN Advanced Institute for Computational Science’s (AICS) K-Computer defending its position in the number-one spot.

The number two and number three spots also remained fixed with the National Supercomputing Center in Wuxi’s Sunway TaihuLight as second and Lawrence Livermore National Laboratory’s DOE/NNSA/LLNL Sequoia in third.

Despite the top three machines leading again, a new machine made its way to the 5th ranking on this year’s list: Leibniz-Rechenzentrum’s SuperMUC-NG.

The SuperMUC-NG hails from Munich and boasts more than 311,040 cores and a peak performance of 26.9 petaflops with Intel Skylake processors.

“The Graph500 is a leading indicator of high-performance computing (HPC) development and often reveals trends regarding new technologies used in HPC machines,” said Georgia Tech School of Computational Science and Engineering Chair David Bader.

“It provides a benchmark standard to test a supercomputer’s abilities to construct, search, and conduct edge-detection for undirected graphs.”

Bader is a founding member of the Graph500 and serves as one of Graph500 executive committee members. He is joined by Peter Kogge of the University of Notre Dame, Andrew Lumsdaine of Indiana University, Richard Murphy of Micron Technology Inc., Torsten Hoefler of ETH Zurich, and Anton Korzh of Micron Technology, Inc.  

This committee – along with an International Multidisciplinary Steering Committee comprising 30 international HPC experts from academia, national laboratories, and industry – rank Graph500 machines based on the benchmark standards.

The Graph500 list was established in 2009 to complement an effort to create a benchmark for high-performance computing efforts. By creating this benchmark, the Graph500 serves as a way to indicate to academia, industry, and government the importance of progression in this field, as well as to maintain interest and engagement from world HPC leaders.