Configuration

Hardware Configuration

• 48 standard compute nodes

  • 4× NVIDIA GH200 Grace-Hopper Superchip (see also node image)

    • CPU: NVIDIA Grace (Arm Neoverse-V2), 72 cores at 3.1 GHz base frequency; 120 GB LPDDR5X memory at 512 GB/s (8532 MHz)

    • GPU: NVIDIA Hopper H100, 132 multiprocessors, 96 GB HBM3 memory at 4 TB/s

    • NVIDIA NVLink-C2C CPU-to-GPU link at 900 GB/s

    • TDP: 680 W (for full GH200 superchip)

  • NVLink 4 GPU-to-GPU link, 300 GB/s between pairs of GPUs (150 GB/s per direction)

  • Network: 4× InfiniBand NDR200 (Connect-X7)

• 1 login nodes

  • 1× NVIDIA GH200 Grace-Hopper Superchip (differences highlighted)

    • CPU: NVIDIA Grace (Arm Neoverse-V2), 72 cores at 3.1 GHz base frequency; 480 GB LPDDR5X memory at 384 GB/s (6400 MHz)

    • GPU: NVIDIA Hopper H100, 132 multiprocessors, 96 GB HBM3 memory at 4 TB/s

    • NVIDIA NVLink-C2C CPU-to-GPU link at 900 GB/s

    • TDP: 900 W (for full GH200 superchip)

  • Infiniband NDR (Connect-X7)

  • 100 Gigabit Ethernet external connection

  • local disk for operating system (1× 960 GB NVME)

• 13824 CPU cores

• 200 Gbit/s network connection to JUST

Node diagram of the 4× NVIDIA GH200 node design of JUPITER Booster / JEDI. Links and bandwidths are added.

Software Overview

• RHEL 9 distribution

• JUPITER Management Stack

  • Eviden Smart Management Center xScale

• Scientific Software (EasyBuild)

  • GCC

  • NVIDIA HPC Compiler (_NVHPC_)

  • OpenMPI

• IBM Storage Scale (GPFS) parallel file system