linux-kernel.vger.kernel.org archive mirror
 help / color / mirror / Atom feed
From: miquels@cistron-office.nl (Miquel van Smoorenburg)
To: linux-kernel@vger.kernel.org
Subject: Re: [khttpd-users] khttpd vs tux
Date: Sat, 3 Nov 2001 18:18:04 +0000 (UTC)	[thread overview]
Message-ID: <9s1ccs$5cd$2@ncc1701.cistron.net> (raw)
In-Reply-To: <20011103162642.A25824@fenrus.demon.nl> <Pine.LNX.4.30.0111031740300.8812-100000@mustard.heime.net>

In article <Pine.LNX.4.30.0111031740300.8812-100000@mustard.heime.net>,
Roy Sigurd Karlsbakk  <roy@karlsbakk.net> wrote:
>> tux is more advanced than khttpd. It's also more intrusive to the kernel as
>> far as core changes are concerned. These changes allow for higher
>> performance, but you'll only notice that if you want to fill a gigabit line
>> or more.....
>
>Are there any good reasons why to run khttpd, then?
>What I need is a server serving something between 50 and 500 concurrent
>clients - each downloading at 4-8Mbps.
>Which one would be best? Anyone have an idea?

Seriously? 500*8 Mbit/sec = 4 Gbit/sec

In that case you need at least 10 boxes, each with a gigabit card,
with loadbalancing through DNS. Each box will do max. 400 mbit/sec
and have 50 clients on it - standard apache will do fine, I think.
Otherwise just add a few boxes.

You will need a Juniper M20 or a Cisco 124xx series with 2xSTM16 (OC64)
or 1x10GigE upload capacity and 10xGigE slots in it. That will cost
as much 100-200 of the Linux boxes so the Linux boxes are the least
of your worries. Not to mention the cost of 4 Gbit/sec of Internet
bandwidth.

Mike.
-- 
"Only two things are infinite, the universe and human stupidity,
 and I'm not sure about the former" -- Albert Einstein.


  reply	other threads:[~2001-11-03 18:18 UTC|newest]

Thread overview: 18+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
     [not found] <20011103162642.A25824@fenrus.demon.nl>
2001-11-03 16:43 ` [khttpd-users] khttpd vs tux Roy Sigurd Karlsbakk
2001-11-03 18:18   ` Miquel van Smoorenburg [this message]
2001-11-03 20:24   ` Dirk Moerenhout
     [not found] <3BE427CD.702@bewegungsmelder.de>
2001-11-03 18:02 ` Roy Sigurd Karlsbakk
2001-11-03 19:08   ` Thomas Lussnig
2001-11-03 19:14     ` Roy Sigurd Karlsbakk
2001-11-03 19:21     ` Alan Cox
2001-11-03 19:18       ` Roy Sigurd Karlsbakk
2001-11-03 19:31         ` J Sloan
2001-11-04  1:19           ` Alan Cox
2001-11-05 10:15           ` Ingo Molnar
2001-11-06  4:46             ` J Sloan
2001-11-03 19:37         ` Alan Cox
2001-11-04  0:07         ` Erik Mouw
2001-11-04 15:32           ` John Alvord
     [not found] <3BE42379.2050604@bewegungsmelder.de>
2001-11-03 17:08 ` Roy Sigurd Karlsbakk
     [not found] <20011103165129.B26040@fenrus.demon.nl>
2001-11-03 16:56 ` Roy Sigurd Karlsbakk
2001-11-03 16:21 Roy Sigurd Karlsbakk
2001-11-04  5:26 ` [khttpd-users] " Chul Lee

Reply instructions:

You may reply publicly to this message via plain-text email
using any one of the following methods:

* Save the following mbox file, import it into your mail client,
  and reply-to-all from there: mbox

  Avoid top-posting and favor interleaved quoting:
  https://en.wikipedia.org/wiki/Posting_style#Interleaved_style

* Reply using the --to, --cc, and --in-reply-to
  switches of git-send-email(1):

  git send-email \
    --in-reply-to='9s1ccs$5cd$2@ncc1701.cistron.net' \
    --to=miquels@cistron-office.nl \
    --cc=linux-kernel@vger.kernel.org \
    /path/to/YOUR_REPLY

  https://kernel.org/pub/software/scm/git/docs/git-send-email.html

* If your mail client supports setting the In-Reply-To header
  via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).