I’m not sure what you mean by that - do you mean 100MBs RAM usage? I explain how that works here.
Unless your pipelines spill to disk or are writing an output file they shouldn’t be that slow. Can you give some examples of pipelines that are slow? If you are using MongoDB version 4.4 then explain with executionStats will show how long each stage took as well as how many documents entered each stage.
Example:
db.coll.explain("executionStats").aggregate( [ <your-pipeline-here> ])
I walk through an example of this at the end of the talk I did for last year’s MDB.live event but unfortunately the resolution of the code didn’t turn out that good…
Asya