What are the two different methods?

Interactive processing

This is the simplest way to work on any system - you simply log on and run whatever commands you need to, whether on the command line or in a graphical environment and you log out when you've finished.

Interactive jobs run directly on the limited number of login nodes (servers) that each cluster has (two on ALICE and six on SPECTRE). Although user login sessions are spread across these nodes, interactive processes will be competing with each other.

Batch processing

This is slightly more difficult as work has to be more carefully planned and able to run without user interaction. Batch jobs are submitted to a job scheduler and run on the first available compute node(s). Once submitted, users can log off and simply wait for the jobs to complete, with the option to be sent an email when this has happened. This means that whilst batch jobs may have to wait in a queue before running they are then given a dedicated set of resources to ensure that, once running, they complete as quickly as possible.

Which method should I use?

Short tasks, those which require frequent user interaction and those which are graphically intensive are the type that are best done interactively. Batch jobs are best for longer running processes, parallel processes or for running large numbers of short jobs simultaneously. If something can be left running for a signficant amount of time without any interaction then it's almost certainly a batch job. Running processes as batch jobs also has the benefit of allowing you to log out of the system and to log out of the PC you're connecting from.

How short is "short"?

This is different between ALICE and SPECTRE.

ALICE

The design of ALICE is strongly biased towards batch jobs having 256 compute nodes compared to 2 login nodes so interactive processes shouldn't take more than a few minutes of CPU time.

SPECTRE

SPECTRE has been designed to allow more interactive use so has 6 login nodes compared to 67 compute nodes. Here, interactive processes may use up to an hour or two of CPU time. There are hard limits applied which means that processes will be killed if it uses more than 2 hours of CPU time. Note that a process which is largely idle may take much longer than 2 hours to consume this amount of CPU time.

What if I need to run longer processes interactively?