httpd-dev mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From Jason Clary <>
Subject Re: ZD Net tests, IRIX, and MAX_SOCKETS
Date Mon, 20 Jan 1997 14:13:41 GMT
> They only tested one server under Unix?  That's lame.  Sure, they can
> claim that they just wanted to test Unix as an alternative to NT so they
> only needed one, etc. but it leaves an awful lot of room for
> misconfiguration... 

Well... one unix OS...  They tested NCSA and Enterprise Server under unix
as well but on the same IRIX box.

They all came out about the same rating.

That CAN'T be right.. ;P

> Are they doing SSL transfers though?  They are real pigs.  It is possible
> (don't know, haven't looked) that the SSL part is slowing it down.

I doubt it, I don't think WebBench does SSL, but I could be wrong.
I'll download a copy today and test that theory, its available from
ZDNet as shareware.  
> If they aren't doing SSL transfers, I agree 100/sec is way too low unless
> they are doing some really odd transfers.

Looks like simple stuff.. they give you a set of test pages to put on
the server with WebBench.. even test CGI's if you want to run a non-static
test (which they didn't for the mag article)

Anyways, it does seem exceptionaly wrong.. especialy considering NCSA,
and Stronghold seem to rate the same..  They managed to get 200 out of
Enterprise for the same OS..  maybe its just faster at tossing individual
files and exiting...  That could be something very simple that would change
the average served file from 2.5 sec to 1 sec...  Just multithreading
would probably fix that.. But its still under 256 max sockets, most likely.

View raw message