Currently our application is undergoing performance testing. We don't have any automated tool to conduct the load test and we are doing it manually with 10 machines executing the command simultaneously. But our average number of concurrent users will be 30 -35 users. Is it comfirtable to say for 10 concurrent users an elapsed time of say x and for 30 concurrent users it will be 3x. Please clarify.
<BLOCKQUOTE><font size="1" face="Verdana, Arial, Helvetica">quote:</font><HR>Originally posted by sivashan: Is it comfirtable to say for 10 concurrent users an elapsed time of say x and for 30 concurrent users it will be 3x. Please clarify. <HR></BLOCKQUOTE>
It is not appropriate to make that extrapolation.
At some point, performance wil degrade significant;y. It might be 15 users, it might be 100 users, but there is always a point where it happens. At some point the system will become unusable.
If you test with 10 users, you know what will happen with 10 users. Nothing more.