Started in 1993, the TOP500 lists the fastest computers used today, ranked according to Linpack benchmark results. Published twice a year, the TOP500 list provides an important tool for tracking usage trends in high-performance computing. The 50th TOP500 List was released during the November Supercomputing’17 conference.

TOP500 50th List Highlights

InfiniBand Accelerates 77 Percent of New High-Performance Computing Systems on TOP500 Supercomputer List. InfiniBand Connects 2 of the Top 5 Supercomputers, Including World’s Fastest Supercomputer. InfiniBand Connects 6 times more new HPC systems versus Proprietary Interconnects (June’17 - Nov’17). Mellanox Connects all 25 Gigabit and above Ethernet Systems.

Highlights of InfiniBand and Mellanox usage on the November 2017 TOP500 list:

  • Mellanox accelerates the fastest supercomputer on the list
  • InfiniBand connects 2 of top 5 systems - #1 (China) and #4 (Japan)
  • InfiniBand connects 6 times more new HPC systems versus proprietary interconnects (June’17 - Nov’17)
  • InfiniBand connects 15 times more new HPC systems versus Ethernet (June’17 - Nov’17)
  • InfiniBand connects 77% of new HPC systems (June’17 – Nov’17)
  • Mellanox connects 39 percent of overall TOP500 systems (192 systems, InfiniBand and Ethernet)
  • InfiniBand connects 33 percent of the total TOP500 systems (164 systems)
  • InfiniBand connects 60 percent of the HPC TOP500 systems
  • 25G Ethernet first appearance on the Nov’17 TOP500 list (China Hyperscale companies) - 19 systems
  • Mellanox connects all of 25G, 40G and 100G Ethernet systems on the Nov’17 TOP500 list
  • InfiniBand is the most used high-speed Interconnect on the TOP500
  • InfiniBand is the preferred interconnect for Artificial Intelligence and Deep Learning systems
  • Mellanox enables highest ROI for Machine Learning, High-Performance, Cloud, Storage, Big Data and more applications

Mellanox ConnectX-5 EDR 100Gb/s InfiniBand adapters, Switch-IB2 based switch systems and LinkX fiber and copper cables optimize server and storage performance and provide the scalable, low-latency, and power-efficient interconnect for the world’s fastest supercomputers.