[Hipsec-rg] About simultaneous mobility to HIP

[gurtov at cs.helsinki.fi (Andrei Gurtov)]

-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA1

Hi,

I suppose there is a danger in splitting the host identifier in several
subdomains that it stops to be an identifier and becomes a locator. This
is exactly like IP address space is allocated, from national to regional
ISP down to company up to the user. What happens e.g. if ISP A has own
host IDs and ISP B own host IDs and a host wants to handover between
providers? Then we have the same problems as currently with IP addresses.

Andrei

xuxiaohu 41208 wrote:
|> I've commented to John in private that the use of DHT or Hi3 (or even
|> fast enough DynDNS) could help in simultaneous mobility.
|
| Hi all,
|
| I have made a presentation of HRA(Hierarchical Routing Architecture)
on the HIPRG session this Monday, but no comment received up to now.
|
| The major differece between HRA and HIP is that HRA uses a
hierarchical and cryptographic hash identifier which is composed of
Administrative Domain (AD) ID and a hash of the ADID and the public key.
It looks like CGA. However, the AD ID, which embeds organization
semantics,is only routable on the ID/locator mapping system which is
realized by Hierarchical DHT. The main purposes are 1)Enforce
organization-level access control policy 2)Overc
| ome the lack of economic&trust model in the id/locator mapping system
for the flat identifier.
|
| I wonder whether the flat HIT is good enough in the DHT or Hi3 overlay
from different perspectives.
|
| Any comment is welcomed.
|
| Xiaohu XU
-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.4.9 (MingW32)
Comment: Using GnuPG with Mozilla - http://enigmail.mozdev.org

iEYEARECAAYFAkiRsHgACgkQP7jp0uceFkQXUQCfYCercVob4mqy2i3bzMPY5G6S
wlQAoLYq0PfoHkfLRRoicxoXidzYlpHT
=fIf3
-----END PGP SIGNATURE-----