Jim Jagielski wrote:
>>On Wed, Sep 05, 2001 at 08:05:27PM -0700, Ian Holsman wrote:
>>
>>>Some performance results with mod_gz are available at
>>>http://webperf.org/a2/v25/
>>>(no core dumps.. pages look ok on a real browser while running test)
>>>I'm going to be re-running the tests for a longer period to see if
>>>there are memory leaks (as well as quantify/purify on it next week)
>>>
>>>brief summary:
>>> * page requests down 16% when using GZ
>>>
>
>I don't get that one... Why would compressing the data in stream
>result in any change in page requests?
>
The rate at which this particular benchmark client issues requests
is throttled by the speed at which the server can issue them. (It
uses a fixed number of client threads, each of which does one blocking
HTTP request at a time.) The results suggest that the time spent
compressing the data exceeded the reduction in transmission time
that the compression yielded. (This probably is to be expected with
a 100 Mb/s data path between the clients and the server.)
--Brian
