Peter, I think, I did outline MY kind of clean-slate solution which is based on topology aggregation and NOT on topology address aggregation (in the shortest email-thread possible). I expressed why I can't display all of the solution at the moment but I was clearly pointing out that the scalability problem is an immanent problem of the DV- and address aggregating concept. I also launched the postal delivery service discussion as to demonstrate this, i.e. that forwarding inside much bigger networks while using another paradigm would never have this problem. Hereby, it seems that my position has been understood. drc wrote, not long ago: To state the obvious, by divorcing the identity from the location, you allow that identity to "easily" change location. Conceptually, the benefits of such a split would include: - end users/sites having multiple providers (multihoming) without having to participate in the locator routing system - end users/sites changing from one provider to another over a long time period (i.e., changing ISPs) - end users/sites changing from one provider to another over short time periods (i.e., mobility, depending on how the mapping is performed) - ISPs being able to rearrange network topology without significantly impacting network users
IMO: Excellent, particularly, if said in view of my solution. Heiner In einer eMail vom 07.07.2008 14:19:07 Westeuropäische Normalzeit schreibt [EMAIL PROTECTED]: Therefore, we can solve the scalability problem through topological address aggregation -IF- we remove -identity- from protocol layer 3. Yes and there is a consensus on that, right? Thanks, Peter --- On Fri, 7/4/08, William Herrin <[EMAIL PROTECTED]> wrote: > From: William Herrin <[EMAIL PROTECTED]> > Subject: Re: [RRG] Long term clean-slate only for the RRG? > To: [EMAIL PROTECTED] > Cc: [EMAIL PROTECTED], [EMAIL PROTECTED], [email protected] > Date: Friday, July 4, 2008, 11:38 AM > On Fri, Jul 4, 2008 at 10:55 AM, Peter Sherbin > <[EMAIL PROTECTED]> wrote: > >> -MY- point was that Line 1 need not be there at > all. It is an > >> identifier which serves no role in the routing. > > > > It sure does as long as there are more than one person > living at the > > same address. The selection does not stop until it > reached > > the "end". This is why defining the end > point is critical. It will help > > with setting all of the identifier properties. > > Not so! Once the letter has reached the address, folks at > the address > are allowed to open the letter and make further decisions > based on > what's inside, handing it to a human being, the trash > can or even back > to the post office with a new address. > > > On Fri, Jul 4, 2008 at 11:04 AM, Marshall Eubanks > <[EMAIL PROTECTED]> wrote: > > It is not clear to me how any of this discussion helps > routing research for > > the Internet. > > Follow the logic chain: > > A. -IF- topological address aggregation was practical, the > route > scalability problem could be readily solved by aggregating > routes > based on the address aggregation. > > B. Topological address aggregation would be practical -IF- > any > endpoint's layer 3 address could be routinely and > recursively > reassigned by the "upstream" routers through an > address assignment > protocol without disrupting layer 4 AND the node could > sensibly handle > multiple defaults with multiple source addresses via a > routing policy > protocol. > > C. An ephemeral address which changes without disturbing > layer 4 would > be possible -IF- the node identity value used by layers 4 > and above > WAS NOT derived from the layer 3 address. In other words, > make layer 4 > treat the layer 3 address the way layer 3 treats the layer > 2 address. > > Therefore, we can solve the scalability problem through > topological > address aggregation -IF- we remove -identity- from protocol > layer 3. > > > So, the relevance of the discussion about the name (line 1) > in a > postal address is this: The name (identity) obviously > isn't needed for > the post office to successfully route the letter. Routing > still works > if your name isn't present on the envelope. If the same > is true of > network packets in a hypothetical architecture (and it > should be) then > we can solve the layer 3 routing problem by changing how > the layer 4 > protocols determine a node's identity. > > After all, I'm not "3005 Crane Drive," > I'm "William Herrin." And the > post office can deliver mail to "3005 Crane > Drive" without knowing > whether it's intended for "William Herrin." > Fix how layer 4 handles > host identity and the layer-3 routing system no longer > needs to manage > a large database. > > Of course, layer 4 now needs to manage a large map from > identities to > their current locations, but we've already seen that > well handled by > (insert drum roll) DNS. > > Regards, > Bill Herrin > > > > -- > William D. Herrin ................ [EMAIL PROTECTED] > [EMAIL PROTECTED] > 3005 Crane Dr. ...................... Web: > <http://bill.herrin.us/> > Falls Church, VA 22042-3004 -- to unsubscribe send a message to [EMAIL PROTECTED] with the word 'unsubscribe' in a single line as the message text body. archive: <http://psg.com/lists/rrg/> & ftp://psg.com/pub/lists/rrg
