Quantcast
Channel: CSCS User Forum » Recent Posts
Browsing all 35 articles
Browse latest View live

Luca Marsella on "File transfer to CSCS machines"

Q: I'm a new user at CSCS. I cannot use ftp to connect to the system and upload files. Could you give some advice?

View Article



Luca Marsella on "File transfer to CSCS machines"

You should be able to use scp or sftp instead: check the availability of these commands using which scp or which sftp on the terminal. For more information on their usage, have a look at the manual...

View Article

Luca Marsella on "How to manage my jobs in the queue?"

Q: How can I list all my jobs in the system, get details on each of them and delete one of my jobs in the queue?

View Article

Luca Marsella on "How to place and release a job from hold state?"

Q: How can I place my SLURM job into an hold state? And how can I release the hold?

View Article

Luca Marsella on "How to have the list of available nodes and partitions?"

Q: How can I view the list of available partitions and nodes and obtain detailed information on them?

View Article


Luca Marsella on "How to display job accounting information?"

Q: How can I display accounting informations of my SLURM jobs?

View Article

Luca Marsella on "Getting information on the configuration and nodes available"

Q: How can I see the detailed configuration of the current SLURM system and see the list of the available nodes?

View Article

Luca Marsella on "What is the meaning of job state codes?"

Q: I have submitted a job with SLURM, but I don't understand the meaning of the different job state codes displayed in output by the command squeue.

View Article


Luca Marsella on "What is the meaning of job state codes?"

Jobs typically pass through several states in the course of their execution. The typical states are PENDING, RUNNING, SUSPENDED, COMPLETING and COMPLETED. An explanation of each state follows:CA...

View Article


Luca Marsella on "Getting information on the configuration and nodes available"

The corresponding commands are scontrol show config and scontrol show nodes respectively.

View Article

Luca Marsella on "How to display job accounting information?"

You can have this information with the command sacct -l -u $USER

View Article

Luca Marsella on "How to have the list of available nodes and partitions?"

The command sinfo -a -l gives you a list of the available partitions.Typing scontrol show partition PARTITION you will get details on the chosen partition: if you don't specify a partition, you will...

View Article

Luca Marsella on "How to manage my jobs in the queue?"

You can see your jobs in the queue issuing the command squeue -a -l -u $USER.In order to see details on a specific job, type scontrol show JOB_ID.If you want to delete one of your submitted jobs, the...

View Article


Luca Marsella on "How to place and release a job from hold state?"

In order to place a job on hold type scontrol hold JOB_ID.To release the job from the hold state, issue scontrol release JOB_ID.`

View Article

Luca Marsella on "Where can I run Matlab, do you have a template script?"

Matlab is available as a module on Julier for post-processing, but you may load the module directly from Ela, switching the default module environment from rosa to julier first:module switch rosa...

View Article


Luca Marsella on "How to chain jobs between Rosa and Julier?"

Q: I would like to run a post-processing job on Julier which depends on the data produced by another job running on Rosa. How can I set such a chain of jobs?

View Article

Luca Marsella on "How to chain jobs between Rosa and Julier?"

Chaining jobs between machines means launching two dependent jobs, with the second one starting after the first one has finished, but on different systems (Rosa and Julier in this case). First you have...

View Article


Luca Marsella on "How to optimize the number of tasks per node?"

Q: My application requires more than 1 GB per task, so I had to reduce the number of tasks per node. Is there an optimal choice for the number of tasks per node on Rosa?

View Article

Luca Marsella on "How to optimize the number of tasks per node?"

Remember that you are charged resources per node: of course when memory becomes an issue, then you need to reduce the number of tasks per node. However you will gain not only the extra memory but also...

View Article

Luca Marsella on "How to transfer efficiently data to /project"

Q: How can I move a considerable amount of data efficiently to my /project folder?

View Article
Browsing all 35 articles
Browse latest View live




Latest Images