How are Users different from concurrent users when doing Performance testing

[Preethi Srinivasan]

The performance testing blog in JIRA, talks about User groups and concurrent users  creation required for performance testing. While concurrent user/ actions impacts performance, can you please explain why we need to create users also in JIRA. How does the number of users registered in JIRA and groups  have an impact in performance testing. Is it not sufficient to create concurrent users using the test tools. I am planning to do performance testing for JIRA (for a medium/large)  environment.

[Nic Brough -Adaptavist- · Accepted solution]

User = an account that Jira is aware of

Concurrent users = people making the server do work at this time

You might have 1,000 users, 500 active and able to log in, but the concurrent users are anonymous people and the number of the 500 logged in - those clicking on stuff/typing/using REST

Every user account, active or not, makes the server do work - you might have one person logged in, they click on an issue and Jira has to go through the list of users it knows of to work out which of them might be displayed in the "assignee" drop-down.  It does not have a huge effect, but a speed test on a Jira with 10 user accounts will give you a different result to a speed test on the same Jira after you've added 4990 more.