You are viewing a plain text version of this content. The canonical link for it is here.
Posted to users@httpd.apache.org by "Wm.A.Stafford" <st...@marine.rutgers.edu> on 2007/03/23 14:43:35 UTC
[users@httpd] Calculating maximum throughput for an Apache based system
I have been asked to come up with a figure for the number of users per
unit time our
Apache-based site can support and to indicate the resource; memory, disk
space etc, that
would be maxed out when this number of users was reached.
There are so many assumptions that would have to be made, such as
average time
per request, average memory usage per request that I don't think such a
measure
would be very realistic. Nevertheless, I have to come up with something.
Are there any existing statistics or algorithms that deal with this sort
of thing?
I apologize in advance for this desperate fishing expedition.
-=bill
---------------------------------------------------------------------
The official User-To-User support forum of the Apache HTTP Server Project.
See <URL:http://httpd.apache.org/userslist.html> for more info.
To unsubscribe, e-mail: users-unsubscribe@httpd.apache.org
" from the digest: users-digest-unsubscribe@httpd.apache.org
For additional commands, e-mail: users-help@httpd.apache.org
Re: [users@httpd] Calculating maximum throughput for an Apache based
system
Posted by Robin-David Hammond <kb...@pins.net>.
I have one stat to offer.
NetBSD 4.0_BETA2 (ANDROMEDA-$Revision: 4.212 $) #0: Wed Jan 17 07:52:19
PST 2007
total memory = 1536 MB
avail memory = 1485 MB
timecounter: Timecounters tick every 10.000 msec
found Grand Central at 0xf3000000
mainbus0 (root)
cpu0 at mainbus0: 750 (Revision 2.3), ID 0 (primary)
cpu0: HID0 8290c2a4<EMCP,ECLK,DOZE,DPM,ICE,DCE,SPD,SGE,BTIC,BHT>,
powersave: 1
cpu0: 500.00 MHz, no-parity 2MB WB L2 cache (FB SRAM) at ratio
.
.
.
pchb0 at pci0 dev 11 function 0
pchb0: Apple Computer Bandit Host-PCI Bridge (rev. 0x03)
esiop0 at pci0 dev 13 function 0: Symbios Logic 53c896 (ultra2-wide scsi)
esiop0: using on-board RAM
esiop0: interrupting at irq 23
scsibus0 at esiop0: 16 targets, 8 luns per target
esiop1 at pci0 dev 13 function 1: Symbios Logic 53c896 (ultra2-wide scsi)
esiop1: using on-board RAM
esiop1: interrupting at irq 23
scsibus1 at esiop1: 16 targets, 8 luns per target
this machine has pushed about 45 Mbps no sweat.
the busses are PCI 32/33
Robin-David Hammond KB3IEN
On Fri, 23 Mar 2007, Wm.A.Stafford wrote:
> I have been asked to come up with a figure for the number of users per unit
> time our
> Apache-based site can support and to indicate the resource; memory, disk
> space etc, that
> would be maxed out when this number of users was reached.
>
> There are so many assumptions that would have to be made, such as average
> time
> per request, average memory usage per request that I don't think such a
> measure
> would be very realistic. Nevertheless, I have to come up with something.
>
> Are there any existing statistics or algorithms that deal with this sort of
> thing?
>
> I apologize in advance for this desperate fishing expedition.
>
> -=bill
>
>
> ---------------------------------------------------------------------
> The official User-To-User support forum of the Apache HTTP Server Project.
> See <URL:http://httpd.apache.org/userslist.html> for more info.
> To unsubscribe, e-mail: users-unsubscribe@httpd.apache.org
> " from the digest: users-digest-unsubscribe@httpd.apache.org
> For additional commands, e-mail: users-help@httpd.apache.org
>
>
---------------------------------------------------------------------
The official User-To-User support forum of the Apache HTTP Server Project.
See <URL:http://httpd.apache.org/userslist.html> for more info.
To unsubscribe, e-mail: users-unsubscribe@httpd.apache.org
" from the digest: users-digest-unsubscribe@httpd.apache.org
For additional commands, e-mail: users-help@httpd.apache.org
Re: [users@httpd] Calculating maximum throughput for an Apache based system
Posted by Joshua Slive <jo...@slive.ca>.
On 3/23/07, Nick Kew <ni...@webthing.com> wrote:
> On Fri, 23 Mar 2007 09:43:35 -0400
> "Wm.A.Stafford" <st...@marine.rutgers.edu> wrote:
>
> > I have been asked to come up with a figure for the number of users
> > per unit time our
> > Apache-based site can support and to indicate the resource; memory,
> > disk space etc, that
> > would be maxed out when this number of users was reached.
>
> We have some folks who have talked or written about scaling Apache
> to absurd amounts of traffic. OTTOMH, Colm MacCarthaigh has reported
> over 100,000 concurrent connections to a single server, and Brian Akins
> is another name to google for. You might like to read what they've
> done to scale it.
That's "Brian Atkins" (just to make the googling a little more fruitful).
But as you seem to partially recognize, this question is
almost-meaningless without discussing a very-specific setup. For
example, Colm hits those numbers for a site that offers large static
files. You'll never approach that for complex php scripts.
So the best advice is not to look for other people's stats, but to
test yourself. Apache offers a couple tools for this: apachebench and
flood.
Joshua.
---------------------------------------------------------------------
The official User-To-User support forum of the Apache HTTP Server Project.
See <URL:http://httpd.apache.org/userslist.html> for more info.
To unsubscribe, e-mail: users-unsubscribe@httpd.apache.org
" from the digest: users-digest-unsubscribe@httpd.apache.org
For additional commands, e-mail: users-help@httpd.apache.org
Re: [users@httpd] Calculating maximum throughput for an Apache
based system
Posted by Nick Kew <ni...@webthing.com>.
On Fri, 23 Mar 2007 09:43:35 -0400
"Wm.A.Stafford" <st...@marine.rutgers.edu> wrote:
> I have been asked to come up with a figure for the number of users
> per unit time our
> Apache-based site can support and to indicate the resource; memory,
> disk space etc, that
> would be maxed out when this number of users was reached.
We have some folks who have talked or written about scaling Apache
to absurd amounts of traffic. OTTOMH, Colm MacCarthaigh has reported
over 100,000 concurrent connections to a single server, and Brian Akins
is another name to google for. You might like to read what they've
done to scale it.
--
Nick Kew
Application Development with Apache - the Apache Modules Book
http://www.apachetutor.org/
---------------------------------------------------------------------
The official User-To-User support forum of the Apache HTTP Server Project.
See <URL:http://httpd.apache.org/userslist.html> for more info.
To unsubscribe, e-mail: users-unsubscribe@httpd.apache.org
" from the digest: users-digest-unsubscribe@httpd.apache.org
For additional commands, e-mail: users-help@httpd.apache.org