This question is in regard to loops in process models as it relates to Appian Health Checks. Recently I have taken on a more administrative role and one of my regular tasks is to review monthly health checks across our environments. During these reviews I see a lot of items related to looping processes. I understand large looped processes may not be performant. The thing that drew my attention and for which I have not found a relevant explanation in documentation is the health check description (see below) the highlighted section says that 10 or more nodes is considered high (makes sense) but 2 or more is medium. That seems like a very low threshold.
1 process flow(s) contain unattended loops with 10 or more nodes (High) and 19 process flow(s) contain unattended loops with 2 or more nodes (Medium). Looping is typically an indication of using processes to perform extract-transform-load (ETL) type operations. It is recommended to use specific ETL tools (such as a database) to implement complex and bulk background operations.
I realize that high and medium do not define an absolute metric but I regularly would design processes that looped over 2-4 nodes and I never considered that a bad or risky practice. Is there something here I'm not understanding?
Discussion posts and replies are publicly visible
You probably already knew this, but as some level of reassurance, just remember that much of their automated guidance needs to be taken with a large grain of salt.
I also consider that threshold very low. I'd be much more concerned about the amount of possible loops or the amount of data possibly being accumulated in PVs than the amount of nodes in a loop.