About the Dardel HPC system
The Swedish National Infrastructure for Computing (SNIC) has invested in a new high-performance computing (HPC) system for Swedish researchers. The new flagship system is an HPE Cray EX supercomputer, called Dardel, that is being installed at PDC during 2021 and 2022.
If you are interested in seeing what is inside the HPE Cray EX supercomputer systems in general, you can explore here .
The new Dardel system at PDC is being installed in two phases. When fully installed, the system will have several partitions, including
- a large CPU partition (that will be suitable for a broad range of computational applications), and
- a partition based on graphics processing units (GPUs), which is intended for the most computationally demanding applications.
Installation of the CPU partition (providing 70,912 CPU cores) and the storage system started in August 2021. This first phase of the system has been available for research since December 2021. The plan is for the GPU partition to be installed mid-2022, and to be available to researchers from the summer of 2022.
Each compute node has two AMD EPYC™ Zen2 2.25 GHz 64 core processors, which means that each compute node has a total of 128 physical CPU cores. Two virtual hardware threads are enabled on each physical CPU core, which means that each compute node has a total of 256 logical/virtual cores.
There are several sizes of compute nodes in the CPU partition.
- 488 × 256 GB (SNIC thin nodes)
- 20 × 512 GB (SNIC large nodes)
- 8 × 1024 GB (SNIC huge nodes)
- 2 × 2048 GB (SNIC giant nodes)
- 36 × 256 GB (KTH industry/business research nodes)
This gives a total of 554 nodes.
This partition will be equipped with 56 GPU nodes, each of which will be configured with:
- one AMD EPYC™ processor with 64 cores (which is currently under development)
- 512 GB memory
- four AMD Instinct™ MI250X GPUs
- HPE Slingshot (ethernet-based) using Dragonfly topology
- 100 Gb/s for the first phase, which will be upgraded to 200 Gb/s in the second phase
The LINPACK performance of the first phase of the system is 2.28 Petaflops. The overall theoretical peak performance of the whole system is expected to increase to about 13.5 Petaflops when the second phase is operational.
- available for academic research via SNIC and PRACE
- available for business and industrial research projects
- operation currently guaranteed to be funded till Q3 2026