Typically concurrency is measured in the number of connections a server or portal is handling at any given time. A possible measurement is also how many users have visited a portal or site over a period of an hour; however the number of connections does not necessarily indicate load. For example: A user may be logged in, but is spending up to 10 seconds reviewing what they are seeing to determine their next course of action resulting in no activity/load. Therefore for this performance test, the users are scripted with appropriate think times of 4-5 seconds or so (to simulate real-time activities and load. If an application identifies end users by process of login and logout procedure then the number of concurrent users for that particular application under performance testing is expected to support this work-flow which contains Login and Logout activity.
What do you mean by concurrent users. If you say 50 concurrent users, do you mean 50 users are hitting the GUI at once, every second, and so we’re looking at 50X60 users per minute. Or is it 50 concurrent users per 5 seconds, etc?
Go back to the interview and tell them you don’t know the answers and they should hire someone that knows what they are doing.
--
You received this message because you are subscribed to the Google "LoadRunner" group.
To post to this group, send email to LR-Loa...@googlegroups.com
To unsubscribe from this group, send email to
LR-LoadRunne...@googlegroups.com
For more options, visit this group at
http://groups.google.com/group/LR-LoadRunner?hl=en
Hi,
It depends on the application and business processes in picture.
The ideal way of creating a performance test scenario is by finding out the exact user activity from the production servers and replicating the same in the load that we are designing.
The ideal way also comprises of including thinktime values that as the value a real user will wait on a particular page for.
Eg. If on www.google.com page a real user waits for 5 seconds(while he is entering the keyword and clicking 'search'), then the think time to be used for this transaction while customizing the script would be 5 seconds.
Hence 50 concurrent users can be 50 users with 5 seconds of think time (or) 50 users with 10 seconds of think time. Idea is the replicate the real time situation exactly the way it is.
I hope I am making sense.
Thanks,
Swaraj
--
Thanks for the nice article
Sent from my iPhone