bounded storage model - why is R organized as 2-d array?

Chris Palmer chris at eff.org
Thu Mar 9 13:37:02 EST 2006


Steven M. Bellovin writes:

> I think you vastly overestimate how much hardware one needs to do
> something like AES.  I ran
> 
>   	dd if=/dev/zero bs=32k count=1024| openssl speed aes-128-cbc
> 
> on a 1500 Mhz Athlon.  It reported speeds of ~27.5 MBps, or 220 Mbps.
> Even video isn't that fast, and that's a slow CPU by today's standards.

Right, but even though a 1.5GHz machine is a bit old (heh...) for a
workstation, my dinky little Linksys WRT54GC wireless AP still needs to
AES-encrypt a theoretical maximum of 54Mbps when I turn on WPA.

http://www.linksysinfo.org/modules.php?name=Content&pa=showpage&pid=12&page=1

Its adorable little CPU is a far cry from a 1.5GHz Athlon. :) I haven't
tested to see if there is any difference in throughput with WPA on vs.
off, but I'd be surprised if there weren't.

Thus, something faster than AES, but still strong, would be nice. Your
point about CPU cache size vs. pad size is well-taken, though.


This message signed and nonrepudiable for your reading pleasure,
-- 
https://www.eff.org/about/staff/#chris_palmer

-------------- next part --------------
A non-text attachment was scrubbed...
Name: not available
Type: application/pgp-signature
Size: 186 bytes
Desc: not available
URL: <http://www.metzdowd.com/pipermail/cryptography/attachments/20060309/ecceb7fd/attachment.pgp>


More information about the cryptography mailing list