News! More nodes will be available for HPC calculations from early 2025.

• Node pez047 (previously ANAV partition and now R630-v4 partition) is opened.
• Nodes pez048 and pez049 (HM and HM-dev partitions) are fully opened.
• Nodes pez051 (R182 and R182-open partitions) and pez052 (R640 partitions) are opened.
• Node pez053 (R6525 partitions) is added.

A stricter policy will be applied regarding job extensions in order to democratize access to Acuario and allow jobs to be submitted more often.
From now on, job extensions are only allowed for up to 5 extra days. In most partitions, the limit is set to 10 days by default. In the event that more calculation time is needed, a request must be made to higher instances.

New node pez053 counts with 256 CPUs, to democratize its use a set of restriccions are applied to its partitions. As a difference to the rest of the cluster, to send jobs to the new partition R6525, now you will need to use this flag:

#!/bin/bash
# Script.sh file
#SBATCH --qos=cpu-limit32

Or this other option:

$ sbatch script.sh --qos=cpu-limit32

This will limit the CPUs permitted to a maximum of 32 CPUs per job. Apart from this extra flag, make sure you don’t overpass this limit or your job won’t run.

As you all can see, the HPC website is renewed for a more modern look and more information has been added about the computing resources available in Acuario. Check Computing resources page to see new nodes resources.

If you need any help, feel free to ask via https://tickets.cimne.upc.edu/

A new node, pez049, has been added to acuario. This node is the same than pez048, but with 1TB of RAM. You can use it via HM partition.

Also, we have upgraded pez048 with double of main memory.

A new node called pez048 has been added to the cluster. Its characteristics has been selected in order to boost shared memory jobs. This node has 2 CPU AMD EPYC 7451, each one with 24 cores (48 threads). Unlike the rest of the nodes, threads are enabled due to every thread owns a FPU. Also, at tests with enabled threads have been proved that there are no bottlenecks. Moreover, the node has an amount of 512 GB of RAM DDR4 at 2.6 GHz.

In order to use it, a new partition called HM has been created. Also, its corresponding development partition (HM-dev) has been created.

If you need to know additional specs about this node feel free to ask for it.

On the other side, B510 partition has been updated with the addition of 4 nodes (pez0[17-18] & pez0[33-34]).

Once nodes pez0[17-32] have been checked, we have created a new nodes partition called B510 with nodes pez0[19-32] (nodes pez0[17-18] will be added soon).

Also, between this week and next one, we will install a new compute node with 512 GB RAM and 2 CPU AMD EPYC 7451 with 48 cores (96 threads) each one.