Thomas,
Close, I believe. The second occurrence of "physical network"
makes
me just a bit queasy. The frame header information is not used by the wires
(or fiber) of the physical network. Nor is it used by other physical network
components that may still be used in some L2 networks (Hubs, repeaters, ...).
802.1D bridges also do not use VLAN IDs to make forwarding
decisions
and 802.1Q bridges operate in 802.1D compatibility mode when not configured
with VLAN information.
How about:
"VLANs are a well understood construct in the networking industry,
providing an L2 service via a physical network in which tenant
forwarding information (i.e. - the VLAN ID) is included in the L2
frame header and used by VLAN bridges in making forwarding
decisions."
Also, the term "tenant" is an alien term in a general VLAN
context,
but I can live with this usage because we clarify what we mean by the term
in the parenthetical text.
Some folks question why this is needed. If this was left out,
what
do we lose? Putting it in context, and leaving the tenant part out, this is
what we would have:
5.3. 802.1 VLANs
VLANs are a well understood construct in the networking industry,
providing an L2 service via an L2 Virtual Network. A VLAN is an L2
bridging construct that provides the semantics of virtual networks
mentioned above: a MAC address can be kept unique within a
VLAN, but it is not necessarily unique across VLANs. Traffic scoped
within a VLAN (including broadcast and multicast traffic) can be kept
within the VLAN it originates from. Traffic forwarded from one VLAN
to another typically involves router (L3) processing. The forwarding
table look up operation may be keyed on {VLAN, MAC address} tuples.
VLANs are a pure L2 bridging construct and VLAN identifiers are
carried along with data frames to allow each forwarding point to know
what VLAN the frame belongs to. Various types of VLANs are available
today, which can be used for network virtualization even together.
The C-VLAN, S-VLAN and B-VLAN IDs are 12 bits. The 24-bit I-SID
allows the support of more than 16 million virtual networks.
Note that the difference between this and the text in section
5.3
now is that phrase "in-band" has been removed. We may be going to a
lot of trouble to explain a concept that isn't needed in the context of the
draft.
--
Eric
-----Original Message-----
From: Thomas Narten [mailto:[email protected]]
Sent: Wednesday, July 03, 2013 8:19 AM
To: Eric Gray
Cc: [email protected]
Subject: Re: [nvo3] IESG review of draft-ietf-nvo3-overlay-problem-statement-03
Importance: High
Eric,
> Thomas,
>
> We really cannot say -
> " VLANs are a well understood construct in the networking
> industry, providing an L2 service via a physical network in
> which tenant forwarding information is part of the [physical]
> network infrastructure."
> - because this is not anything that VLANs are "well understood" to do.
> The term "tenant" is alien in the context of VLANs, and the (tenant
> associated) information is not part of the physical network, but is
> (instead) included as part of the frame-header for Ethernet frames
> forwarded in the context of each VLAN.
Yep, I messed up on the rewording a bit here.
> Perhaps if I knew why you were proposing this change to the earlier
> text, I could provide a better formulation?
To address Adrian's comment and observation that "in-band" was barely used and
arguably didn't need to be in the terminology section...
> I think we are trying to do is characterize the way that the L2
> service is provided by VLANs is "in-band" – by which we apparently
> mean that the mechanism used for VLANs is to carry identifiers for
> virtual networks in the data-frames themselves, as part of the
> Ethernet frame-header.
Right. And a key difference between "in-band" and overlays is that in the
in-band case, the network infrastructure uses those identifiers while making
forwarding decisions. With overlays, the identifiers are only used at the edge
and not in the forwarding plane (at least not directly -- deep packet
inspection muddies this, but is really just an optimization).
> This might be part of the changes you indicated you would do to
> address Adrian's comment. If that is the case, I would suggest that
> the text should be changed to:
> "VLANs are a well understood construct in the networking industry,
> providing an L2 service to VLAN member end-stations where VLAN
> forwarding information is included in the PDU header of each data
> PDU."
This is better, but would prefer something like:
VLANs are a well understood construct in the networking industry,
providing an L2 service via a physical network in which tenant
forwarding information (i.e., the VLAN ID) is included in the L2
frame header and used by the physical network while making
forwarding decisions.
(this avoids using "PDU", which the PS document hasn't used...)
Thomas
_______________________________________________
nvo3 mailing list
[email protected]
https://www.ietf.org/mailman/listinfo/nvo3
