RE: What is the standardization status of AES in IPSec?

["Andrew Wenlang Zhu" <Andrew_zhu@hp.com>]

I think it a good idea to begin with one key length such as 128. Because
this key length can fit in the current IKE and virtually needs no dramatic
changes. Larger DH group MODP length may be the only exception.

Actually, if IETF do not make 128 entropy a MUST for AES-128, DH 1024 and
1536 MODP is still a viable thing. One of the benefits of using AES in this
way is its performance. In our HP lab available soft-implementation of AES
and 3DES, AES only takes almost 1/3 of 3DES time in en/de-crypt ion.
Performance is crucial especially to software-only IPSec implementation.
Interesting enough, I found nearly the same speed ratio in some hardware
implementation.

Andrew
-----------------------------
Andrew Zhu
HP Systems Networking Solution Lab
 IP Security & System Firewall Project
 Andrew_zhu@hp.com


>-----Original Message-----
>From: owner-ipsec@lists.tislabs.com
>[mailto:owner-ipsec@lists.tislabs.com]On Behalf Of
>steve.robinson@metaware.com
>Sent: Friday, February 01, 2002 1:32 PM
>To: Scott Fanning
>Cc: Andrew Wenlang Zhu; ipsec@lists.tislabs.com;
>owner-ipsec@lists.tislabs.com
>Subject: Re: What is the standardization status of AES in IPSec?
>
>
>
>Well, since FIPS-197 says that one of the three key lengths MUST be
>supported, and that the other two SHOULD be supported, it's
>probably a good
>idea to follow that lead, and choose a single key size that must be
>supported, and the keep the others optional.
>
>
>
>
>
>                     "Scott Fanning"
>
>                     <sfanning@cisco.com       To:     "Andrew
>Wenlang Zhu" <Andrew_zhu@hp.com>,
>                     >
><ipsec@lists.tislabs.com>
>
>                     Sent by:                  cc:
>
>                     owner-ipsec@lists.t       Subject:
>Re: What is the standardization status of AES in IPSec?
>                     islabs.com
>
>
>
>
>
>                     02/01/02 02:59 PM
>
>
>
>
>
>
>
>
>
>and on that note.. If AES is the MUST implement algorithm, does that
>include
>all key sizes?
>
>Scott
>----- Original Message -----
>From: "Andrew Wenlang Zhu" <Andrew_zhu@hp.com>
>To: <ipsec@lists.tislabs.com>
>Sent: Friday, February 01, 2002 11:45 AM
>Subject: What is the standardization status of AES in IPSec?
>
>
> > Hello:
> >
> > Can any one give me an update on the standardization status
>of using AES
>in
> > IPSec?
> >
> > I am reading "The AES Cipher Algorithm and Its Use With IPsec"
> > <draft-ietf-ipsec-ciph-aes-cbc-03.txt> and read " Once NIST
>has published
> > the AES FIPS ... AES should become a default and
>mandatory-to-implement
> > cipher algorithm for IPSec".
> >
> > FIPS-197 was out in Nov-2001. When an IPSec/AES RFC is
>expected to come
>out?
> >
> > Thanks,
> > ---------------------------------------
> > Andrew Zhu
> > HP Systems Networking Solution Lab
> > IP Security & System Firewall Project
> > Andrew_zhu@hp.com
> >
> >
>
>
>
>
>