Re: [Apn] should add the gap analysis for GENEVE (RFC8926)

[Gyan Mishra <[e-mail]>]

Hi Shuping

This is in similar context to use of VXLAN-GPE to carry APN APP ID marking
information

https://datatracker.ietf.org/doc/html/draft-ietf-nvo3-vxlan-gpe-11


   The capabilities of the VXLAN-GPE protocol can be extended by
   defining next protocol "shim" headers that are used to implement new
   data plane functions.  For example, Group-Based Policy (GBP) or In-
   situ Operations, Administration, and Maintenance (IOAM) metadata
   functionalities can be added as specified in
   [I-D.lemon-vxlan-lisp-gpe-gbp
<https://datatracker.ietf.org/doc/html/draft-ietf-nvo3-vxlan-gpe-11#ref-I-D.lemon-vxlan-lisp-gpe-gbp>]
and
   [I-D.brockners-ippm-ioam-vxlan-gpe
<https://datatracker.ietf.org/doc/html/draft-ietf-nvo3-vxlan-gpe-11#ref-I-D.brockners-ippm-ioam-vxlan-gpe>].


GENEVE

https://tools.ietf.org/html/rfc8926

Work such as "VL2: A Scalable and Flexible Data Center Network" [VL2
<https://tools.ietf.org/html/rfc8926#ref-VL2>]
   and "NVO3 Data Plane Requirements" [NVO3-DATAPLANE
<https://tools.ietf.org/html/rfc8926#ref-NVO3-DATAPLANE>] have
described
   some of the properties that the data plane must have to support
   network virtualization.  However, one additional defining requirement
   is the need to carry metadata (e.g., system state) along with the
   packet data; example use cases of metadata are noted below.  The use
   of some metadata is certainly not a foreign concept -- nearly all
   protocols used for network virtualization have at least 24 bits of
   identifier space as a way to partition between tenants.  This is
   often described as overcoming the limits of 12-bit VLANs; when seen
   in that context or any context where it is a true tenant identifier,
   16 million possible entries is a large number.  However, the reality
   is that the metadata is not exclusively used to identify tenants, and
   encoding other information quickly starts to crowd the space.  In
   fact, when compared to the tags used to exchange metadata between
   line cards on a chassis switch, 24-bit identifiers start to look
   quite small.  There are nearly endless uses for this metadata,
   ranging from storing input port identifiers for simple security
   policies to sending service-based context for advanced middlebox
   applications that terminate and re-encapsulate Geneve traffic.



You can see some similarities between the two NVO3 overlays GENEVE and
VXLAN-GPE both having the ability carry metadata and so would be perfect
for DC environment to carry APN ID in the metadata field for the endpoint
characteristics signaling to the network.

Kind Regards


Gyan

On Tue, Mar 23, 2021 at 10:26 PM Pengshuping (Peng Shuping) <
[e-mail]> wrote:

> Thank you, Linda!
>
>
>
> When we start exploring the solution using GENEVE, we would need to know
> more about the use cases. Thank you for the information!
>
>
>
> BR,
>
> Shuping
>
>
>
> *From:* Linda Dunbar [mailto:[e-mail]]
> *Sent:* Tuesday, March 23, 2021 11:16 AM
> *To:* Pengshuping (Peng Shuping) <[e-mail]>;
> [e-mail]; [e-mail]
> *Subject:* RE: should add the gap analysis for GENEVE (RFC8926)
>
>
>
> Shuping,
>
>
>
> Here is one example: for 5G Edge Computing, the edge devices have limited
> capacity. It can use GENEVE to carry information about the characteristics
> of the App, such as Types, Edge device information, etc.
>
> Linda
>
>
>
>
>
> *From:* Pengshuping (Peng Shuping) <[e-mail]>
> *Sent:* Sunday, March 21, 2021 8:39 PM
> *To:* Linda Dunbar <[e-mail]>;
> [e-mail]; [e-mail]
> *Subject:* RE: should add the gap analysis for GENEVE (RFC8926)
>
>
>
> Hi Linda,
>
>
>
> I was wondering about the concrete usage scenarios since I am not familiar
> with those with GENEVE. For example, in what scenario carrying what
> information to do what?
>
>
>
> Any references on the IoT case you mentioned about?
>
>
>
> Thank you!
>
>
>
> Best regards,
>
> Shuping
>
>
>
> *From:* Linda Dunbar [mailto:[e-mail]
> <[e-mail]>]
> *Sent:* Saturday, March 20, 2021 11:31 AM
> *To:* Pengshuping (Peng Shuping) <[e-mail]>;
> [e-mail]; [e-mail]
> *Subject:* RE: should add the gap analysis for GENEVE (RFC8926)
>
>
>
> ShuPing,
>
>
>
> GENEVE is to carry metadata associated with the packet. Metadata can be
> location information, compute information, service ID, App category, etc..
>
> One open source project even used the GENEVE to replace HTTP for IoT
> devices that don’t support Web browser to communicate with other devices.
>
>
>
> Linda
>
>
>
> *From:* Pengshuping (Peng Shuping) <[e-mail]>
> *Sent:* Friday, March 19, 2021 9:11 PM
> *To:* Linda Dunbar <[e-mail]>;
> [e-mail]; [e-mail]
> *Subject:* RE: should add the gap analysis for GENEVE (RFC8926)
>
>
>
> Hi Linda,
>
>
>
> Yes, this is what I meant.
>
>
>
> The draft-li-6man-app-aware-ipv6-network-02 gives an example of the
> encapsulation of the APN attribute in the IPv6 data plane.
>
>
>
> Since you mentioned GENEVE, have you seen some scenarios that could use
> APN in GENEVE?
>
>
>
> Thank you!
>
>
>
> Best Regards,
>
> Shuping
>
>
>
> *From:* Linda Dunbar [mailto:[e-mail]
> <[e-mail]>]
> *Sent:* Friday, March 19, 2021 11:48 PM
> *To:* Pengshuping (Peng Shuping) <[e-mail]>;
> [e-mail]; [e-mail]
> *Subject:* RE: should add the gap analysis for GENEVE (RFC8926)
>
>
>
> ShuPing,
>
>
>
> When you say “needs to be developed”, do you mean mapping the APN-ID and
> APN-Associated-Parameters into the GENEVE TLV?
>
>
>
> For example, the App-AWARE ID from the
> draft-li-6man-app-aware-ipv6-network-02 can be specified as a Sub-TLV to
> GENEVE header.
>
>
>
> Linda
>
>
>
> *From:* Pengshuping (Peng Shuping) <[e-mail]>
> *Sent:* Friday, March 19, 2021 3:24 AM
> *To:* Linda Dunbar <[e-mail]>;
> [e-mail]; [e-mail]
> *Subject:* RE: should add the gap analysis for GENEVE (RFC8926)
>
>
>
> Hi Linda,
>
>
>
> Thank you!
>
>
>
> GENEVE can be taken into account when carrying the APN attribute. The
> encapsulation with it seems more to be a solution that needs to be
> developed rather than a gap analysis to be added. How do you think?
>
>
>
> Best regards,
>
> Shuping
>
>
>
> *From:* Linda Dunbar [mailto:[e-mail]
> <[e-mail]>]
> *Sent:* Friday, March 19, 2021 7:56 AM
> *To:* [e-mail]; [e-mail]
> *Subject:* should add the gap analysis for GENEVE (RFC8926)
>
>
>
> Shuping, et al,
>
>
>
> I think that the draft-peng-apn-scope-gap-analysis should add the gap
> analysis for GENEVE (RFC8926).
>
>
>
> GENEVE has been used very widely among Cloud networks. For example, you
> can have a GENEVE encapsulated flows to Firewalls. You can attach App
> related Meta data to the GENEVE header, so that the nodes along the way can
> do the needed processing.
>
>
>
>
>
>
>
> Geneve Header:
>
>       +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+
>
>       |Ver|  Opt Len  |O|C|    Rsvd.  |          Protocol Type        |
>
>       +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+
>
>       |        Virtual Network Identifier (VNI)       |    Reserved   |
>
>       +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+
>
>       |                                                               |
>
>       ~                    Variable-Length Options                    ~
>
>       |                                                               |
>
>       +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+
>
>
>
> If there are reasons that GENEVE can’t be used, it is better to document
> those reasons in the Gap analysis.
>
>
>
> Linda Dunbar
> --
> Apn mailing list
> [e-mail]
> https://www.ietf.org/mailman/listinfo/apn
>
-- 

<http://www.verizon.com/>

*Gyan Mishra*

*Network Solutions A**rchitect *

*Email [e-mail] <[e-mail]>*



*M 301 502-1347*