Cost and traffic control vs. Max. user actions per minute exceeded

[Babar_Qayyum]

Dear All,

I am posting this to have a little more clarity in my understanding.

Cost and traffic control reduce the granularity of user-action and user-session analysis by capturing a lower percentage of user sessions.

Is this applicable to the user sessions and user actions count or percentage? for example analyze 30% of user sessions will also be applicable to the user action count/percentage.

Max. user actions per minute exceeded: Dynatrace automatically throttles the number of user sessions being captured until the number of user actions is below the limit again. The throttling is done on a user session basis in order to provide statistically correct values in Dynatrace. If the overall traffic volume drops, Dynatrace will automatically return to the capturing rate you've set for your environment.

During the event period, Dynatrace reduces the amount of captured user sessions on the cluster side. All count metrics are impacted (for example, user session count, user action count, and conversion count).

Regards,

Babar

[ChadTurner]

From my understanding, setting a capture rate for cost control will only affect the number of user sessions captured, not the number of actions within a session.

Basic Rum capture (without replay) is charged when a captured user performs 2 actions on a page. From there you use a slice of licensing. Now that also falls into your license with Dynatrace. Back when it was billed as user sessions, it was a 1 for 1, So one 1 would consume 1 User Session License from your purchased allotment.  

Then Dynatrace Switched to DEM Units, with then stated that each user session, without session replay consumed 1/4th of a DDU. Thus giving you 4 Sessions for 1 DDU. With Session replay added in that was a 1 for 1. 

Now with Dynatrace Moving to DDUs that's another story and something you might need to review with your license to confirm what is included. DDUs made it so you had a pool of DDU's that can be used for a whole mix of things, Synthetics, User sessions and so on. So if you lowered the frequency of your synthetics, you would then be able to use the DDU savings from Synthetics over on the user session aspect. 

So ultimately, when you set a detection threshold, to 50% of users, Dynatrace will grab a random 50, of your 100 Users and record and retain that RUM Data. It will not, grab 50% of users then only provide 50% of the users journey. 

[Babar_Qayyum]

Hello @ChadTurner 

Thank you for your thoughts. The following statement is from the documentation.

Cost and traffic control reduce the granularity of user-action and user-session analysis by capturing a lower percentage of user sessions.

Therefore, I wanted to reconfirm that the cost and traffic control are going to apply to the user-action as well or not.

Regards,

Babar

[ChadTurner]

yeah I think they might be lumping things in, if you reduce the number of user sessions then there for the number of user actions is reduced as well. I can see where it can also be interpreted that user actions would be reduced as well. Maybe Dynatrace can clarify that. 

It is my understanding that setting cost limits only reduces the number of user sessions that are captured via RUM, which would have an affect on your overall user actions as there will be a decline since only 50 of those users are having their sessions monitored and tracked. 

[Babar_Qayyum]

Hello @ChadTurner 

Might be someone else can also shed on it or the Dynatrace will have to update the documentation. I got confused while reading the cost and traffic control with the combination of Max. user actions per minute exceeded.

Regards,

Babar