The new Galileo Infrastructure
We are pleased to announce that the new DGX - Deep learning GPUs Accelerators are available for ISCRA projects. These nodes are equipped with DGX-A100 cards, each one hosting 8 Nvidia A100 GPUs; 2AMD Rome 7742 Processors (128) and 1TB of RAM. They are specifically designed for Machine learning projects, but also suitable for many other projects. DGX cards are in preproduction phase from January 2021.
our Tier-1 system, Galileo, will be updated and improved in the following months. The procedure will start in a month, a pre-production phase is expected for May, the full production in July 2021.
The new Infrastructure is co-funded by the European ICEI (Interactive Computing e-Infrastructure) project, it is a system engineered by DELL and consists of:
- 528 computing nodes 2 CPU Intel CascadeLake 8260, with 24 cores each, 2.4 GHz, 384 GB RAM and 36 additional nodes with 2x NVIDIA GPU V100 with 100Gbs Infiniband interconnection
- 77 computing server OpenStack for cloud computing, 2x CPU 8260 Intel CascadeLake, 24 cores, 2.4 GHz, 768 GB RAM, with 100Gbs Ethernet interconnection.
- 20 PB of active storage accessible from both cloud and HPC nodes.
- 5 PB of fast storage for HPC system.
- 1 PB Ceph storage for Cloud (full NVMe/SSD)
More
Some
technical details about the new system
are available in our UserGuidewill be available soon.