For discussion,
Although some workflows may include runtime hints or suggestions as to test data, the "real" execution may still be quite hard to predict.
One idea that we'd like to bounce around is the idea of essentially allowing users to voluntarily write-back (and the server aggregating) runtime information such as how long a workflow may take to run, how much memory or CPU it took.
This may eventually allow us to collect the equivalent of "system requirements" for a workflow using real-world data.
This may tread on ground that is covered by https://osf.io/h59uh/ (can this form of data be extracted from it?)
┆Issue is synchronized with this Jira Story
┆Project Name: Zzz-ARCHIVE GA4GH tool-registry-service
┆Issue Number: TRS-59
For discussion, Although some workflows may include runtime hints or suggestions as to test data, the "real" execution may still be quite hard to predict. One idea that we'd like to bounce around is the idea of essentially allowing users to voluntarily write-back (and the server aggregating) runtime information such as how long a workflow may take to run, how much memory or CPU it took.
This may eventually allow us to collect the equivalent of "system requirements" for a workflow using real-world data. This may tread on ground that is covered by https://osf.io/h59uh/ (can this form of data be extracted from it?)
┆Issue is synchronized with this Jira Story ┆Project Name: Zzz-ARCHIVE GA4GH tool-registry-service ┆Issue Number: TRS-59