Page 1 of 1

The "Lite" version of the eValid load test facility

PostPosted: Tue Jul 24, 2012 3:54 pm
by sLarry
Does the "Lite" version of the eValid load test facility simulate actual load?

Re: The "Lite" version of the eValid load test facility

PostPosted: Fri Jul 27, 2012 11:10 am
by eValid
sLarry wrote:Does the "Lite" version of the eValid load test facility simulate actual load?
The eVlite LoadTest playback option drives load on the server machines by making simple HTTP/S "Get" requests based on a sequence of URLs, or based on the URLs that appear in a full-reality script.

This imposes load on the server machines by requiring them to deliver more --and in most cases MANY more -- URLs that a regular eValid playback does. And that may be useful to know.

(In practice, we oftn think of eVlite as having the ability to create a lot of busy work, which often is useful to create a burdened system while you run realistic tests in the foreground for the response time data you REALLY want.)

However, you should recognize the difference between such "load" and the requirements placed on a server when it is respond to a full-realism simulation of a user action sequence, which is what a full eValid browser playback does.

The testing/performance engineer who handles your LoadTest work -- which we think of as a continuing set of "experiments" to find out things about how your server responds -- should take this into account. eVlite can be an effective part of a performance testing experiment. Or it can have no effect.

It all depends on your web server, your application, the sophistication you require from your performance testing efforts.

(Sorry, we know you may have expected a definitive answer, but we think it is best in this User Forum to give you the truth as we see it.)

eValid LoadTest Team

You