[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Re: scope of solution..



IPv6 does have good aggregation.  But if we adopt a where and who approach
and reduce the size_t (rtr_entry_point) we cut that size in half.  If we
push out some of the scalability issues to end-nodes we gain even more. If
we find we need to improve compression we gain more.

IPv6 is not the issue.  Once this economy bounces back and the weak .com
players who hyped up our industry are dead only the strong will still
exist the telcos, ISPs, and IXs will have absorbed wireless better and
then lots of new nodes and Internet connectivity will appear.  There will
be way to many addresses needed to deal with no matter what they are.
We are talking billions.  IPv6 gives us the opportunity to distribute the
problem on the network in addition to large address space.  I see no other
way.

/jim

On Wed, 11 Apr 2001, Bora Akyol wrote:

> Bill,
> 
> 
> At billions of prefixes, the routers will not be able to do the lookup in
> a reasonable time anyway so your packets will get there sometime in the
> next century.
> 
> How many unique prefixes do we expect to see in the Internet
> with IPv6? I thought one of the reasons why IPv6 addresses were designed
> the way they are was for ease of aggregation. If it turns out that IPv6
> causes the amount of unique prefixes in the forwarding (not routing)
> tables to blow up to multiple millions or billions then we are in trouble.
> 
> Bora
> 
> 
> 
> 
> On Wed, 11 Apr 2001, Bill Sommerfeld wrote:
> 
> > > Coming back to my initial statement, I have seen routers handle upwards of
> > > 500,000 prefixes with rich PATH attributes already, and as
> > > implementations get more efficient (and distributed) I believe BGP can be
> > > made to scale (depending on the router hw that you choose to buy).
> >
> > How long does this 500000 entry routing table take to converge after
> > something major flaps?
> >
> >Will this approach allow every small business and every household to
> > be multihomed, scaling to billions of prefixes?
> >
> > 					- Bill
> >
> 
>