>Howard,
>I seem to remember @home trying to base their network on RFC 1918
>addressing, and then petitioning for a Class "A" after they ran into
>problems with users that were trying to play games across the Internet. It
>seems that many games rely on a unique IP address to identify users. Does
>this sound familiar to you?
>
>Irwin
Multiplayer games, and some streaming applications, are MAJOR
problems for NAT. They tend to have special-purpose multicast
protocol functions, and they need to know addresses in order to
rendezvous.
The @home, and the cable industry in general, is a bit more complex.
As you know, carriers justify address space based on actual customers
and 3-month projections. The cable industry asked for a huge block,
and the addressing authorities said they didn't have customers.
In response, the industry said that they had millions of real
customers, just not IP customers. They did, however, have the theory
"build it and they will come."
On appeal, the decision was to grant a specific block for the cable
industry, 24.0.0.0 (I think it initially was a /14), so the
utilization could be tracked. At present, the ARIN assumption is
that 7 percent of cable customers will use IP service. ARIN will
grant more than 7 percent if usage can be documented.
>
>-----Original Message-----
>From: Howard C. Berkowitz [mailto:[EMAIL PROTECTED]]
>Sent: Wednesday, June 07, 2000 12:04 PM
>To: [EMAIL PROTECTED]
>Subject: Re: NAT for ISPs
>
>
>Current IP address allocation from ARIN assumes that ISPs will use
>private addressing and NAT for single-homed customers. They are
>reasonable about exceptions, such as protocols that won't work
>through NAT, but the exceptions need to be justified for the ISP to
>continue getting address space.
>
>If the customer is multihomed to more than one ISP, the requirements
>for coordinating NATs among them are accepted to be impossible.
>Multihoming alone, however, does not justify provider-independent
>address space.
>
> >I am a newbie
> >
> >I would hazard a guess that, for an ISP(assuming thousands of users) to do
> >address translation would be a large technical feat, consuming more
>Hardware
> >and other resources than the savings on registered IP addresses could
> >justify
> >
> >Please correct me if I am wrong
> >
> >Tayta
> ><[EMAIL PROTECTED]> schrieb im Newsbeitrag
> >[EMAIL PROTECTED]">news:[EMAIL PROTECTED]...
> > > Dear All,
> > >
> > > Anybody here who work for ISP that use NATed/PATed address for their
> > > clients? Is it appropriate for ISP to use NAT/PAT?
> > >
> > >
> > > Thanks
> > >
> > > Reden
>
>___________________________________
>UPDATED Posting Guidelines: http://www.groupstudy.com/list/guide.html
>FAQ, list archives, and subscription info: http://www.groupstudy.com
>Report misconduct and Nondisclosure violations to [EMAIL PROTECTED]
___________________________________
UPDATED Posting Guidelines: http://www.groupstudy.com/list/guide.html
FAQ, list archives, and subscription info: http://www.groupstudy.com
Report misconduct and Nondisclosure violations to [EMAIL PROTECTED]
