to my knowledge, there isn't a default. I would advise against making all the HTTP sampler get the embedded resource. I mainly work on large sites and performance is usually a hard requirement. The applications that I've worked on first hand used a dedicated image server. In fact, if you look at any website that supports moderate to heavy load, all the images are served off a dedicated image server.
hosting the images on a dedicated server will easily improve the performance of a website by 2-3x depending on how many images each page has. the more images a page has, the greater the performance improvement. hope that helps peter On 5/25/05, Chad La Joie <[EMAIL PROTECTED]> wrote: > Yep, I saw the check box, and I do check it. My thinking was that > instead of having to check that box for each request (and we have a fair > number of them in our test plans) it would be nice if I could have that > be the default behavior, either because it was the default behavior for > the HTTP sampler, or because I was able to set on the HTTP request defaults. > > Peter Lin wrote: > > http://jakarta.apache.org/jmeter/usermanual/component_reference.html#HTTP_Request > > > > if you look at the Http request, you should see "retrieve all embedded > > resource". if you check that, Jmeter will retrieve all the images. > > Keep in mind that most users are not going to download all the images > > on every single page. Any image that has already been downloaded and > > cached in the browser won't be re-downloaded. the exception to this > > case is when a user explicitly sets the browser to always retrieve > > every image. > > > > I'm assuming there's access logs from production. What I tend to do is > > look at the log report and figure out the ratio of images to pages. > > this way, you get closer to simulating real traffic conditions. > > > > peter > > > > On 5/25/05, Chad La Joie <[EMAIL PROTECTED]> wrote: > > > >>When load testing web applications you generally want to simulate, as > >>closely as possible, what would happen if a real user was doing their > >>thing. So, for us, that means that for each HTTP request we want it to > >>retrieve all the resources for that page as well. It seems like this > >>might be a very common thing, perhaps even the normal behavior people > >>wanted. So I was wondering if either the default behavior of the HTTP > >>Request sampler could be to fetch those things, or in a more > >>configurable manner, if there could be an option of the Http Request > >>Defaults config element to set this. > >> > >>Just a thought. > >>-- > >>Chad La Joie 315Q St. Mary's Hall > >>Project Sentinel 202.687.0124 > >> > >>--------------------------------------------------------------------- > >>To unsubscribe, e-mail: [EMAIL PROTECTED] > >>For additional commands, e-mail: [EMAIL PROTECTED] > >> > >> > > > > > > --------------------------------------------------------------------- > > To unsubscribe, e-mail: [EMAIL PROTECTED] > > For additional commands, e-mail: [EMAIL PROTECTED] > > > > -- > Chad La Joie 315Q St. Mary's Hall > Project Sentinel 202.687.0124 > > --------------------------------------------------------------------- > To unsubscribe, e-mail: [EMAIL PROTECTED] > For additional commands, e-mail: [EMAIL PROTECTED] > > --------------------------------------------------------------------- To unsubscribe, e-mail: [EMAIL PROTECTED] For additional commands, e-mail: [EMAIL PROTECTED]
