Answer Posted / krish..
i will explain u with same above example...
MIN MAX AVG 90%ILE
2SEC 9SEC 5 SEC 8SEC
the above what is your result. Fine.. now answer to my questions, we are running 100 users,
1)50 users took only 2 secs, so MIN response time is 2 sec
2)1 users took 9 secs, so MAX response time is 9 Secs
3)Remaining users took some where 6 to 7 secs.
Now as per the above results you cant pick the MIN value 2Sec and say this is the response time for 100 users, in the same way you cant pic MAX value 9secs and say this is response time for 100 users, and u cant take avg 5secs also because 50 users means more half of the users took only 2 secs.
So we need to take the 90% value. Average means 50% value we are not supposed to take it, 90% value will give us a time assuring that this is the most maximum time given to any user who will get response back from the server.
hope you understood this now...
| Is This Answer Correct ? | 0 Yes | 2 No |
Post New Answer View All Answers
How can we perform functional testing under load?
In a web server, database and network are all working fine and now, where could be the problem?
How can performance bottlenecks be identified?
What is protocol function?
What is lr-function?
How you edit the script?
Suppose ur app server is .net server. In that connection pool is related to memory ?explain how?
How can we develop the database vuser script?
When a scenario run exactly what happened?
What is gui vuser and on which platform it will run?
how to get current system time Loadrunner function?
What is the loadrunner start-transaction and its syntax?
What is transaction per second graph(pass)?
How you set maximum number of vuser that a host can run?
How can correlation be performed?