[phpslash-users] include (benchmarks)

[Ajay S. <ss...@od...>]

> IF I wasn't using PHPSlash in a 'production environment' I'd
> volunteer to test it out, and see how much actual difference it
> makes.  I think it's a cool idea - someone want to test it by doing
> some hand-joining of the current tree?

yeah, I just did that this morning (I really should do work when I'm at
W-O-R-K), the results below...

> (And DAMN would it make it simple to install - one script file to
> install, plus the templates, then edit the config file?  Try and top
> that for ease :-)

well, you'd still have the driver scripts like storyAdmin and topicAdmin
so you're not out of the woods yet...

Here's the output of the ab script WITH the AllClasses.php

Server Software:        Apache/1.3.17
Server Hostname:        ...
Server Port:            80

Document Path:          /
Document Length:        31639 bytes

Concurrency Level:      16
Time taken for tests:   152.122 seconds
Complete requests:      200
Failed requests:        0
Total transferred:      6367800 bytes
HTML transferred:       6327800 bytes
Requests per second:    1.31
Transfer rate:          41.86 kb/s received

Connnection Times (ms)
              min   avg   max
Connect:        0    48   918
Processing:  2392  9202 50098
Total:       2392  9250 51016



And here's the output without:



Server Software:        Apache/1.3.17
Server Port:            80

Document Path:          /
Document Length:        31639 bytes

Concurrency Level:      16
Time taken for tests:   160.930 seconds
Complete requests:      200
Failed requests:        0
Total transferred:      6367800 bytes
HTML transferred:       6327800 bytes
Requests per second:    1.24
Transfer rate:          39.57 kb/s received

Connnection Times (ms)
              min   avg   max
Connect:        0    57  2161
Processing:  1083  9180 99142
Total:       1083  9237 101303

As you can see, the Requests per Second is a little higher and there's a
higher Transfer Rate which is cool.

This is the first time I used the 'ab' program that comes with apache so
I'm not sure if I ran it right.  I used 'ab -n 200 -c 16 http...'.  My
server was running at some pretty high loads during the benchmark and
I'm sure I could've pushed it higher until it failed but seeing that
this is a real server for other sites I decided not to.

Now these results aren't very scientific because like I said, the server
was in use for other site so the load probably wasn't consistant
between the two runs.

later,
ajay

----------------------------------------------------------------
Satyajot (Ajay) Sharma
ss...@od...
Digital Odyssey System Administrator
"Did you love this world, and did this world not love you?" Grandaddy
----------------------------------------------------------------