Hi folks, is there an appropriate way to change a ...
# configs
r
Hi folks, is there an appropriate way to change a config of a process locally? I.e. I don’t want to change the base.config but I am having issues with a dataset that is requiring more memory than normal for a process. Where would be the right place to change the config of a particular job when I run the pipeline? Is it okay to add it to the config file where I am specifying the parameters?
a
You can supply an extra config file on the cli with
-c
r
and would that have preference over the
base.config
?
p
!resourcehelp
s
You likely need to configure the pipeline to your available computational resources to you machine or infrastructure. Please see here: https://nf-co.re/docs/usage/configuration#max-resources and the following 'Tuning workflow resources' sections of the documentation for more information.
a
I think it should but not 100% sure. You can use a more fine-grained process selector to ensure it would get precedence though I think. So something like:
Copy code
process {
    withName NF_CORE:WORKFLOW:PROCESS {
     cpus = 100
    }
}
r
Thank you both! Always learning new things here 🙂