[ale] Java Applet Load Testing

[Daniel Kahn Gillmor]

On Tue 2008-02-19 15:37:16 -0500, Greg Freemyer wrote:

> Good idea.  Had not thought about that.  I've done IP aliasing, but
> never with a large volume of IPs is there a limit to the number?
> 10,000 would definately handle everything I can envision.

I dunno if there's a hard limit within the kernel.  It can take a
pretty long time to set up 10000 addresses, but nothing seems to
choke, and once they're up, they're up (i think).

On an otherwise unloaded single-processor AMD K6 350Mhz (yes, my test
hardware is very old):

0 squash:~# ip addr flush dev lan
0 squash:~# ip addr show lan | grep inet | wc -l
0
0 squash:~# time for byteA in $(seq 1 100); do for byteB in $(seq 101 200); do ip addr add 10.13.$byteA.$byteB/16 dev lan; done; done

real	2m12.536s
user	0m34.990s
sys	1m30.598s
0 squash:~# ip addr show lan | grep inet | wc -l
10000
0 squash:~# 

w00t!

   --dkg

PS i've never tried to actually use a system configured in this way.
   If you do it, please report back about how it works!
-------------- next part --------------
A non-text attachment was scrubbed...
Name: not available
Type: application/pgp-signature
Size: 826 bytes
Desc: not available
Url : http://mail.ale.org/pipermail/ale/attachments/20080219/df4ab5a9/attachment.bin