Re: [Grinder-use] "Mean time to resolve host" ?

[Travis B. <tra...@ya...>]

bump

--- On Mon, 3/1/10, Travis Bear <tra...@ya...> wrote:

From: Travis Bear <tra...@ya...>
Subject: [Grinder-use] "Mean time to resolve host"  ?
To: "grinder-use" <gri...@li...>
Date: Monday, March 1, 2010, 12:08 PM

I don't understand the behavior Grinder 3.2 is reporting to me about the mean time to resolve the host.  In a single test, I'm seeing a wide range of host resolution times, with values ranging from 0.3 seconds to 11.8 seconds.  The pages that have the highest overall response times (form posts) also have the highest resolve host times.

I have a couple of assumptions that do not all seem to be correct:

 * resolve host time does not include any of the time the server spends processing the request
 * hostname resolution is done at the OS level, not inside Java or The Grinder
 * since the OS resolves host names, resolution should be uniformly slow or fast for all HTTP requests

Any insight on what may be going on
 here?

Thanks!


-Travis



      
-----Inline Attachment Follows-----

------------------------------------------------------------------------------
Download Intel® Parallel Studio Eval
Try the new software tools for yourself. Speed compiling, find bugs
proactively, and fine-tune applications for parallel performance.
See why Intel Parallel Studio got high marks during beta.
http://p.sf.net/sfu/intel-sw-dev
-----Inline Attachment Follows-----

_______________________________________________
grinder-use mailing list
gri...@li...
https://lists.sourceforge.net/lists/listinfo/grinder-use