LEONARDO: new job local storage resource management

Dear Users, As on the other CINECA clusters, also on LEONARDO a temporary storage area local to nodes is available for the job’s execution, accessible via the environment variable $TMPDIR. Differently […]

Leonardo: new $FAST area in production

Dear Users, we are happy to announce that a new all-flash NVMe storage area is now available to all users. The storage is organized as it is for the work […]

Leonardo: slowdown in the $SCRATCH area – UPDATE

Dear Users, the I/O performances on the $SCRATCH area are partially restored but additional operations are required to recover the optimal functionality of the filesystem. We encourage to keep using […]

Leonardo: issue in job submission to booster partition and in cluster access

Dear Users, the maintenance started and the DCGP partition was switched down. However, we are currently experiencing an issue which prevents the submission of jobs on the entire cluster, including […]

Leonardo: temporary disconnection today at 2 pm

Dear Users, due to an urgent operation on Leonardo external network, today at 2 pm one of the connections to the cluster login nodes will be shut down. The cluster […]

Leonardo scheduled maintenance, February 26th-29th

Dear Users, This is to inform you that, due to scheduled extraordinary maintenance operations on Leonardo’s power system: The operations will start at 08:00 a.m. on the 26th, and we […]

Leonardo: DCGP issue on submission solved

Dear Users, we fixed the issue preventing the submission of jobs requesting more than 32 cpus per node on the DCGP partition. Best regards, HPC User Support @ CINECA

Problem on HPC newsletter

Dear Users, you may have received a mail from the HPC News Center saying that you have been unsubscribed. We are working to fix the issue, please ignore future similar […]

Leonardo back in production

Dear users,Leonardo is back in production. We apologize for the unexpected extended duration of the stop. Best regards,HPC User Support – CINECA

Leonardo: unscheduled short stop tomorrow, January 25th

Dear All,due to the need of urgently updating the booster nodes firmware and the board management software, an unscheduled short stop of the cluster will take place tomorrow morning from […]