Running 100 browsers to generate load

Use and application of the eValid server loading (LoadTest) capability. And in the cloud computing context for monitoring and loading.

Running 100 browsers to generate load

Postby Armandoj » Fri Dec 07, 2012 1:43 pm

In running 100 browsers to generate load, aren't you afraid that all you're doing is measuring your local channel capacity?
Armandoj
 
Posts: 1
Joined: Fri Dec 07, 2012 1:26 pm

Re: Running 100 browsers to generate load

Postby eValid » Mon Dec 10, 2012 6:31 pm

We have found that the channel capacity even with 100 eValid instances running in parallel is very seldom "used up."

The percentage of channel usage does vary quite a bit...but even with very heavy loads on the driver and very large pages to download...we rarely seem more than 10% of available channel bandwidth used up.

In fact, even though eValid instances do consume a lot of memory, running 100+ of them on a user account on a machine does not really tax the input/outpu capacity of the machine. When we do 1,000 BUs using a cloud-based machine image even then the I/O channel percentages stay below 5%.

So, while your point is a fair one, experience shows that the loading approach is valid up to 100 BUs per machine and beyond.

eValid Load Team
eValid
 
Posts: 2392
Joined: Tue Jan 01, 2008 12:48 pm
Location: USA


Return to Performance/Load Testing

Design Downloaded from free phpBB templates | free website templates | Free Web Buttons