GPGPU Computing Servers

These servers are intended for GPU computing (GPGPU), focused on computationally intensive tasks, machine learning, data processing, and scientific simulation that require acceleration by graphics hardware.

Any researcher from the center can request access to these servers. Access is granted upon request and validation.

• Non-cluster HPC computing servers: HPC Computing Cluster
• Servers in CESGA: Request access

Access to these servers is restricted to a specific group, specific project, or is more tightly controlled due to resource management and planning issues.

It is essential to check the updated information in Xici at the time of requesting the service, which details the particular circumstances of each server (access criteria, priorities, usage conditions, etc.).

Not all servers are available at all times for any use. To access the servers, it must be requested in advance through the incident report form. Users who do not have access permission will receive a message indicating incorrect password.

To connect to the servers, you must do it via SSH. The names and IP addresses of the servers are as follows:

The connection is only available from the center's network. To connect from other locations or from the RAI network, you must use the VPN or the SSH gateway.

On servers where there is a Slurm queue manager, its use is mandatory for submitting jobs to avoid conflicts between processes, as two jobs should not run at the same time.

To submit a job to the queue, the srun command is used:

srun programa_cuda argumentos_programa_cuda

The srun process waits for the job to execute to return control to the user. If you do not want to wait, session managers like screen can be used, allowing the job to run in the background and disconnect the session without worrying and retrieve the console output later.

Alternatively, nohup can be used by sending the job to the background with &. In this case, the output is saved in the nohup.out file:

nohup srun programa_cuda argumentos_programa_cuda &

To view the status of the queue, the squeue command is used. The command shows output similar to this:

JOBID PARTITION     NAME     USER  ST       TIME  NODES NODELIST(REASON)
9  servidore ca_water pablo.qu    PD       0:00      1 (Resources)
10 servidore ca_water pablo.qu    PD       0:00      1 (Priority)
11 servidore ca_water pablo.qu    PD       0:00      1 (Priority)
12 servidore ca_water pablo.qu    PD       0:00      1 (Priority)
13 servidore ca_water pablo.qu    PD       0:00      1 (Priority)
14 servidore ca_water pablo.qu    PD       0:00      1 (Priority)
 8 servidore ca_water pablo.qu     R       0:11      1 ctgpgpu2

An interactive view, updated every second, can also be obtained with the smap command:

smap -i 1

On servers using ts as the job manager, it is mandatory to use it to run tasks that use the GPU, in order to avoid conflicts and ensure correct resource allocation.

To request a GPU, the option -G 1 (or the number of GPUs needed) must be added:

ts -G 1 programa_cuda argumentos_programa_cuda

For example:

ts -G 1 python train.py --epochs 100

The system will handle putting the job in the queue and executing it when a GPU becomes available.

To consult more advanced examples (multiple GPUs, additional resources, specific options, etc.), the command can be used:

usage-overview