Closed load model contradicting results?

[Martijn Kuypers]

We've been working with closed workload models with gatling

But we noticed some odd behavior when comparing the full simulation output with the generated report.

When we define 1 concurrent user the simulation output also shows one concurrent user, however when we look at the graphs we noticed the amount of "active users" runs up to 20 users,

Do we need to interpret active users differently from concurrent users? 

Simulation log: http://snippi.com/s/yo0wpym (external snippet because google groups crashed on me trying to post it here)

Screenshot of report 

Thanks in advance!

[Stéphane LANDELLE]

Please check what "active users" are: https://gatling.io/docs/current/general/reports/#active-users-over-time

Those are NOT concurrent users.

Stéphane Landelle

GatlingCorp CTO

slan...@gatling.io

--
You received this message because you are subscribed to the Google Groups "Gatling User Group" group.
To unsubscribe from this group and stop receiving emails from it, send an email to gatling+u...@googlegroups.com.
For more options, visit https://groups.google.com/d/optout.

[Martijn Kuypers]

Thank you for clearing that up.

this raises another question for me, why is the amount of active users interesting for us? 

[slan...@gatling.io]

This metric comes from the first days of Gatling where closed workload model wasn’t officially supported but could be kind of implemented as a hack.

We needed a metric that could show user activity and that would work for both open and closed workload models, given the limited room we have in the HTML reports.

Gatling 3 introduced official closed workload model support, but we could only add arrival rate and concurrent users metrics in FrontLine where we have all the more we want to add more charts.

[Martijn Kuypers]

Thank you for the info!