RE: vpls scaling question

Ping Pan <[email protected]>
Newsgroups gmane.ietf.ppvpn
Message-ID <[email protected]>
On Mon, 12 May 2003, Joris wils wrote:

> Steve,
>
> Questions for you:
> a) Why does each household need its own VLAN?  Why can a set of household
> not share a vlan, each household with its own, possibly statically assigned,
> MAC addresses?
>

Wonder the same thing.

IMHO, each household can rely on L3 routing for connectivity. L2VPN is a
critical element at IP backbone and transport networks only.

> b) If each household must have its own VLAN, then are such VLANs usually
> point2point Ethernet circuits to an Internet router?  It is far easier to
> handle very large numbers of point2point circuits, than very large numbers
> of multipoint VLANs.
>

Follow up question, do we need point-to-multiple VLAN in the backbone?

- Ping

> Cheers,
> Joris
>
> -----Original Message-----
> From: Wright, Steven [mailto:[email protected]]
> Sent: Monday, May 12, 2003 1:21 PM
> To: 'Loa Andersson'
> Cc: '[email protected]'
> Subject: RE: vpls scaling question
>
>
> Loa,
>
> 1) I'm not trying to create requirements here, just to understand what the
> group consensus is. i.e. is this  vpls thing intended to emulate current
> switch capabilities/constraints or do something more ?
>
> 2) Perhaps my terminology was confusing. I was referring to the case of a
> service provider with 10,000,000  customers, each with their own service
> instance. This indeed would be a blessing ! (assuming they all pay their
> bills). Each customer may only want a relatively few VLANs...
>
> 3) Is this a realistic target application ?? well that's always debatable!
> But consider the case of 1 AS per LATA -  and what's the max number of
> households / LATA ? Well there are metro's with 1-10,000,000 households...
>
> 4) Specific deployments depend on a lot more than design constraints (e.g.
> $).....
>   so ...see 1) above...
>
> regards
> Steven Wright
>
> > -----Original Message-----
> > From: Loa Andersson [mailto:[email protected]]
> > Sent: Sunday, May 11, 2003 6:17 AM
> > To: [email protected]
> > Cc: [email protected]
> > Subject: Re: vpls scaling question
> >
> >
> > Steve,
> >
> > just to make sure we are on the same field. For me a vpls is
> > a service I
> > as an oprator offer a customer, could you please give me a scenario
> > where that customer would need 10,000,000 VLANs.
> > If you are talikng about number of vpls's in the operator network,
> > 10,000,000 "service instances" is not a problem, in fact it would be a
> > blessing :)
> >
> > /Loa
> >
> > Steven.Wright wrote:
> > > to clarify the INTRA- AS vpls case
> > > I would like to better understand the expected capabilities
> > of  INTRA-AS
> > > vpls vs a network of Ethernet switches.
> > >
> > > What is the scaling objective for vpls ?
> > > Current generation Ethernet switches have evolved from the 4096
> > > VLAN/network to 4096 VLAN/port. Scale limits on  SP
> > Ethernet networks
> > > are typically driven by other factors e.g. MAC address table size,
> > > issues with multicast etc.
> > >
> > > Is vpls intended to tackle any of those scale constraints ?
> > or to put it
> > > another way...
> > > Current Ethernet switch networks can support on the order
> > of 10,000 VLAN
> > > service instances.  Will vpls enable scaling to the
> > 10,000,000 or so
> > > service instances/AS necessary to enable Ethernet as a
> > residential service ?
> > >
> > >
> > >
> > >
> > >
> > >     -----Original Message-----
> > >     *From:* Steven.Wright [mailto:[email protected]]
> > >     *Sent:* Wednesday, April 23, 2003 5:04 PM
> > >     *To:* *Subject:* vpls scaling question
> > >
> > >     vpls is proposed as a mechanism to enable large-scale
> > Ethernet networks.
> > >     Can anyone quantify for me at what stage flat Ethernet networks
> > >     break and vpls is required ?
> > >     i.e what should the decision criteria be to convert an Ethernet
> > >     network to a vpls network ?
> > >     Steven Wright
> > >     BellSouth
> > >
> >
> >
> > --
> > /Loa
> >
> > mobile + 46 739 81 21 64
> > email: [email protected]
> >
>
>
> *****
> "The information transmitted is intended only for the person or entity to
> which it is addressed and may contain confidential, proprietary, and/or
> privileged material.  Any review, retransmission, dissemination or other use
> of, or taking of any action in reliance upon, this information by persons or
> entities other than the intended recipient is prohibited.  If you received
> this in error, please contact the sender and delete the material from all
> computers."
>
>
>
lmpx.com only provides a reader for public news (NNTP) servers. It is not affiliated with the servers or forums shown here and is not responsible for the content of articles, which is written by their respective authors.