Opened 9 years ago

Closed 9 years ago

#86 closed technical (fixed)

On the design goals and requirements (comment 10 reported by Y. Rekhter)

Reported by: wmhaddad@… Owned by:
Priority: major Component: draft-ietf-lisp
Severity: - Keywords:
Cc:

Description

Section 2

Some of the design goals of this proposal include:

  1. Require no hardware or software changes to end-systems (hosts).
  1. Minimize required changes to Internet infrastructure.
  1. Be incrementally deployable.
  1. Require no router hardware changes.
  1. Minimize the number of routers which have to be modified. In

particular, most customer site routers and no core routers
require changes.

  1. Minimize router software changes in those routers which are

affected.

  1. Avoid or minimize packet loss when EID-to-RLOC mappings need to

be performed.

Wrt (2), (5), and (6) above, how could one determine whether this proposal really minimizes required changes?

Wrt (4), if this statement means existing routers could be software upgraded to be LISP routers, this statement is questionable, for a number of reasons, covered elsewhere in this critique. These include control plane bandwidth (existing routers typically have much lower "punt" processing bandwidth than hardware switching bandwidth) and
the complex LISP forwarding plane, which may not be implementable in microcoded forwarding engines, especially those that are older (but still widely deployed).

Wrt claim in (5) that "no core routers require changes", that is incorrect, as section 5 (p.16) states that:

If the assumption proves true and transit networks with links
limited to 1500 byte MTUs are corner cases, it would seem more
cost-effective to either upgrade or modify the equipment in those
transit networks to support larger MTUs or to use existing
mechanisms for accommodating packets that are too large.

Also, the notion that "no core routers require change(s)" is questionable unless one defines the proxies in [INTERWORK] as being non-core routers.

Wrt (7) above, how could one determine whether this proposal really minimizes packet loss? What needs to be documented is how LISP compares with the current routing model in terms of minimizing packet loss.

Change History (2)

comment:1 Changed 9 years ago by yakov@…

  • Resolution set to fixed
  • Status changed from new to resolved

comment:2 Changed 9 years ago by luigi@…

  • Status changed from resolved to closed
Note: See TracTickets for help on using tickets.