You are viewing a plain text version of this content. The canonical link for it is here.
Posted to dev@httpd.apache.org by Daniel Veillard <ve...@rufus.w3.org> on 1998/01/06 22:26:28 UTC
Re: performance comparison
> Are there any numbers about the performance one could expect from
> Apache? Things like hits/second and sustained transfer rate.
>
> I know that these numbers depend highly on operating system and
> hardware, but I´m not looking for exact numbers, just something to
> compare my measurements to and get an idea if I have room for
> optimization left.
Running a robot on a machine on the same Ethernet hub
(server is a biPPro-200 + 256 MB ram + linux 2.0.33, client is a P200
32 MB of ram and linux 2.0.33) apache-1.2.4
The robot is sucking a mirror of the linux documentation project pages,
and is running in a serialized way (one page after another, no concurrency).
Apache + HTTP/1.0
-----------------
1581 pages fetched in 36 seconds
Apache + HTTP/1.1
-----------------
1508 pages fetched in 20 seconds
the HTTP/1.1 test culminated at approx 800 Ko/s. The generated load on
the server was approx 0.3 .
This is not a "perfect" nor "realistic" test (if such a bench could ever
been made :-) , but may give a point of comparison.
Daniel
--
Daniel.Veillard@w3.org | MIT/LCS Room NE43-344 | Today's Bookmarks :
Tel : +1 617 253 5884 | 545 Technology Square | Linux, WWW, Java,
Fax : +1 617 258 5999 | Cambridge, MA 02139 USA | badminton, Kaffe,
http://www.w3.org/People/W3Cpeople.html#Veillard | HTTP-NG and Amaya.
Re: performance comparison
Posted by Dean Gaudet <dg...@arctic.org>.
If you could give that a whirl with 1.3b3 it'd be great... 1.3b3 should
show an improvement on static sites. Even for HTTP/1.1 pipelining clients
you should see an improvement... unless you've hit your network bandwidth
already :) Oh wait, 800kb/s eh? Well I guess you're just about at limit
anyhow.
I can just picture the code now going crazy assembling writev()s and
mmap()ing files. I should download the most recent version of your robot
someday and make sure 1.3 is assembling writev()s properly for pipelined
clients. Can you send me a link?
Dean
On Tue, 6 Jan 1998, Daniel Veillard wrote:
> Running a robot on a machine on the same Ethernet hub
> (server is a biPPro-200 + 256 MB ram + linux 2.0.33, client is a P200
> 32 MB of ram and linux 2.0.33) apache-1.2.4
>
> The robot is sucking a mirror of the linux documentation project pages,
> and is running in a serialized way (one page after another, no concurrency).
>
> Apache + HTTP/1.0
> -----------------
> 1581 pages fetched in 36 seconds
>
> Apache + HTTP/1.1
> -----------------
> 1508 pages fetched in 20 seconds
>
> the HTTP/1.1 test culminated at approx 800 Ko/s. The generated load on
> the server was approx 0.3 .
>
> This is not a "perfect" nor "realistic" test (if such a bench could ever
> been made :-) , but may give a point of comparison.
>
> Daniel
>
> --
> Daniel.Veillard@w3.org | MIT/LCS Room NE43-344 | Today's Bookmarks :
> Tel : +1 617 253 5884 | 545 Technology Square | Linux, WWW, Java,
> Fax : +1 617 258 5999 | Cambridge, MA 02139 USA | badminton, Kaffe,
> http://www.w3.org/People/W3Cpeople.html#Veillard | HTTP-NG and Amaya.
>
>