Tasks per Jobs as a KPI?

[jblanchett]

Does anyone set target zones for task to job ratios? I'm wondering if this is a useful metric to track. On one side, if jobs > tasks, that means that many jobs are running, but performing no work. On the other hand, a high tasks/jobs ratio could indicate monolithic recipes, but are more likely to indicate high volume jobs processing individual records within a for each loop. Therefore, I'm thinking the higher the tasks/job the better, as long as you can address monolithic recipes somehow. 

Curious to hear other folks thoughts!

[mppowe · Accepted solution]

In a review with Workato, they said they like to see jobs using less than 10 tasks per execution as a general rule of thumb.  That's also assuming that these recipes are trigger-based and being executed frequently throughout the day.  Obviously, if this is a batch job that just runs once per day, that guideline would change.  I'd like to get the point where we have a sense for how many tasks we're consuming from each recipe or project, then as new integration requests come in we could estimate the task usage and make a decision based on the cost/benefit analysis and how many tasks we have remaining.

But... we're not even close to that.  We've had some conversations with Workato about some tooling in that area.

[jblanchett]

Thanks @mppowe ! 10 seems low, but maybe a good guideline for "steps" - not including for each loop executions. 

[mroldanvega]

10 seems very low to me as well. 

@jblanchett how are you mining the data? As part of an ongoing project I am trying to identify recipes that have jobs but have less than normal (or none) tasks, so I was just curious.