Re: Correct way to execute same command in parallel on multiple machines?

[Ole Tange]

On Sat, Feb 3, 2018 at 10:05 PM, Emil Fihlman <emil.fihlman@gmail.com> wrote:
> At the same time the documentation says:
>
> --jobs N
>        -j N
>        --max-procs N
>        -P N     Number of jobslots on each machine. Run up to N jobs in
> parallel.
>                 0 means as many as possible. Default is 100% which will run
> one
>                 job per CPU core on each machine.

This is when you do not use --nonall. When using --nonall the
documentation for --nonall takes precedence.

> So it should by default run a single job on each machine.

By default it should determine the number of cores on your local
machine and run that many jobs in parallel.

So on an 8 core machine it will log into 8 machines in parallel.

I still think what you are looking for is '-j0'.


/Ole