Response rate - Activator on Eventstream vs. Activator on querysets/real-time dashboard

[ObungiNiels]

Hi Fabricators, 

I am experimenting with activators at the moment and have noticed that there seems to be a difference between an activator rule set on an eventstream and one set on a queryset or a dashboard. 

For eventstreams, the activator is set in real-time and actions are also triggered as such. I am currently using this to trigger ETL pipelines based on storage events of an ADLS. 

For querysets/real-time dashboards, I have to set a frequency in which the query is being executed. The shortest frequency is 5 mins. 

From my understanding, this rules out the possibly to trigger "real-time actions" from queryset or reports, due to the rule being checked in a fixed intervall rather than continuously. 

Is my assumption correct that the only way to trigger real-time responses to data in motion is by setting activator rules on an eventstream directly and that subsequent rules on KQL query will suffer from a delay due to the mentioned frequency which has to be set during creation?

Thanks for your input! 

[v-pnaroju-msft]

Thankyou, @tayloramy for your response.

Hi ObungiNiels,

We appreciate your inquiry through the Microsoft Fabric Community Forum.

We would like to inquire whether have you got the chance to check the solution provided by @tayloramy to resolve the issue. We hope the information provided helps to clear the query. Should you have any further queries, kindly feel free to contact the Microsoft Fabric community.

Thank you.

[tayloramy · Accepted solution]

Hi @ObungiNiels, 

I believe your assumption is correct. Activator on a KQL database is limited to run a query to check for the event at 5 minute intervals max. If these queries were being run every second, the CU usage would go through the roof and even F1024 or 2048 capacities would struggle with large datasets. 

In my enviornment, I have two destinations in my eventstreams, one going directly to Activator, and the other going to my KQL database to store the data. THis has been the best approach I've found so far. 

If you found this helpful, consider giving some Kudos. If I answered your question or solved your problem, mark this post as the solution.