New GALILEO in production again

Dear Users,

we are glad to inform you that starting from today, April the 4-th, GALILEO is again available to production.

As a summary, the new configuration of GALILEO (from now on, New GALILEO) is as follows:

• 360 general purpose compute nodes, with 2*18-core Intel(R) Xeon(R) E5-2697 v4 @ 2.30GHz and 128 GB RAM (Broadwell)
• 39 nodes with 2*8-core Intel(R) Xeon(R) CPU E5-2630 v3 @ 2.40GHz (Haswell) + 2*nVidia K80 GPUs
• 1 node with 2*8-core Intel(R) Xeon(R) CPU E5-2630 v3 @ 2.40GHz (Haswell) + 2*nVidia V100 GPUs (for testing purpose only)

The login nodes of the cluster can be accessed via ssh at the address login.galileo.cineca.it.

New GALILEO cluster workload is managed by SLURM workload manager at release 17.11.

New features have been added to the module environment: we kindly invite you to refer to the GALILEO User Guide to check the difference.

For more information about submitting jobs using SLURM, please refer to our simplified Guide on SLURM at https://wiki.u-gov.it/confluence/display/SCAIUS/SLURM+workload+manager

For more information on New GALILEO please refer to the updated User Guide: https://wiki.u-gov.it/confluence/display/SCAIUS/UG3.3%3A+GALILEO+UserGuide

For additional information, please contact superc@cineca.it

Best regards,

HPC UserSupport @ cineca