- Two login nodes
- A Spectrum Scale GPFS parallel file system (380 Tio of usable space, IOs rate : 9 GB/s)
- Intel OPA network 100 Gbit/s
- 192 compute nodes comprising 2 Intel Xeon Gold 6230 20 cores @ 2.1 GHz (Cascade Lake) and 192 GB of RAM
- 14 compute nodes comprising 4 Intel Xeon Gold 6230 20 cores @ 2.1 GHz (Cascade Lake) with 1.5 TB of RAM and 480 GB of SSD disk
- One server comprising 4 Intel Xeon CPU 6148 80 cores @ 2.4 GHz (Skylake) and 1.5 TB of RAM
- 3 servers comprising 4 Intel Xeon CPU 6148 80 cores @ 2.4 GHz (Skylake) and 768 GB of RAM
- 16 compute nodes comprising 2 Intel Xeon CPU 6148 40 cores @ 2.40 GHz (Skylake) and 192 GB of RAM
- 10 GPGPU nodes comprising 2 Intel Xeon Gold 6230 20 cores @ 2.1 GHz (Cascade Lake), 768 GB of RAM and 4 NVIDIA Tesla V100 PCIe graphic cards with 32 GB of GRAM
- 5 GPGPU nodes comprising 2 Intel Xeon Gold 6230 20 cores @ 2.1 GHz (Ice Lake), 1024 GB of RAM and 4 NVIDIA Tesla A100 NVLink graphic cards with 40 GB of GRAM
- 4 GPGPU nodes comprising 2 Intel Xeon CPU 6126 24 cores @ 2.60 GHz (Skylake), 192 GB of RAM and 2 Nvidia Tesla P100 graphic cards with 12 GB of GRAM
- 3 visualization nodes comprising 2 Intel Xeon Gold 6230 20 cores @ 2.1GHz (Cascade Lake), 256 GB of RAM and 1 Nvidia Quadro RTX 5000 graphic card
Ruche (Lenovo supercomputer)
See also this page
User resources
Parallel distributed memory computing
Parallel shared memory and sequential computing
Resources from Fusion (our previous HPE supercomputer)
GPGPU computing
Resources from Fusion (our previous HPE supercomputer)
Visualization
