Arvados

Use Cases:

What if there was a parameter you can set where you can only run N jobs in a pipeline, rather than running the whole thing? What if you already have some of the files down the line in the pipeline and you don't/can't run the first few steps (because you don't have that data [i.e. having bam files already and don't need fastq])? What if you don't need to do some QC and you don't want to waste resources on doing jobs D,E..?

Ideas:

If there was some sort of built in switch in arvados that said you would like to only run N jobs, or the jobs in this branch of this pipeline? It would save computation resources for the bioinformatician if they don't want to run a week long branch of their pipeline.

I think the current workaround would just be to copy the pipeline template and remove the jobs you don't need, but then you could have a bunch of pipeline templates cluttered in your workspace.