Hi Lizhong,

Please see my reply inling.

From: Lizhong Jin [mailto:[email protected]]
Sent: Friday, September 06, 2013 3:19 AM
To: [email protected]; [email protected]; 
[email protected]; Lucy yong; 
[email protected]
Subject: Re: [nvo3] comments on draft-sridharan-virtualization-nvgre-03

We can have a better solution to support ECMP. Most switches and routers today 
support 5 tuple based load balance. Five tuple are IP src/dst addr, tcp|udp 
src/dst ports, and IP protocol type. draft-yong-tsvwg-gre-in-udp-encap-01 
proposes the gre-in-udp encapsulation for GRE encapsulated protocols to be 
tunneled over IP networks where ECMP exists. This solution supports 16 bits 
flow entropy, does not require any change on intermediate switches and routers, 
and applies to any GRE encapsulated protocol. It also gives the ingress end 
point flexibility to generate the flow entropy without explicitly exposing 
VSID. I highly recommend NVGRE proposal to adopt this method for ECMP support.
[PG] - Thanks, I would take a look at the proposal, however I treat this as 
another tunneling protocol in addition to three existing ones i.e. NVGRE, VxLAN 
and STT. What are the specific advantages of this over other protocols? Also 
many of current generation NIC hardware support offloads for the existing three 
protocols. Wouldn't using GRE-IN-UDP break those offloads?
[Lucy] If NIC hardware support VxLAN which use UDP port for the flow entropy, 
it should be easy to support gre-in-udp encaps as well. Will be great  to see 
chip vendor comment on this. Please take a look at the gre-in-udp encapsulation 
proposal (below), welcome comment on it.
[PG] Sure, I would take a look at it.
[Lizhong] it is not easy to design gre-in-udp from the chip design point of 
view. The offloading NIC provided is to terminate tunnel. But in gre-in-udp, 
you have two tunnel layer, then you have to terminate twice for the chip. Then 
of cource, you will increase the cost, power consumption, and etc.
[Lucy] If NIC can terminate UDP tunnel for vxlan and GRE tunnel for NVGRE, it 
can support gre-in-udp for NVGRE. I don't see the given reasons superior to the 
benefit gained in the network.

My concern is if applying gre-in-udp for nvgre, then why not directly using 
vxlan?
[Lucy] I am glad that you point out this.  We also propose vxlan enhancement to 
add prototype field (see 
http://datatracker.ietf.org/doc/draft-yong-l3vpn-nvgre-vxlan-encap/).  These 
enhancements make nvgre+ and vxlan+ very similar semantics and general to 
support other overlays. The only difference in between will be that they use 
different UDP port number and use different bit (3rd bit or 5th ) to indicate 
the presence of VSID or VNI. In the spirit of IETF and for easy interworking, 
it will be nice to have one standardized. Given the existing deployment 
circumstance and chips support both, should IETF compliance and adopt both? 
This is what the community has to decide.

Thanks,
Lucy

Regards
Lizhong


Regards,
Lucy

Here is the draft. The TSVWG will adopt it as WG draft soon.

http://datatracker.ietf.org/doc/draft-yong-tsvwg-gre-in-udp-encap/

Regards,
Lucy


_______________________________________________
nvo3 mailing list
[email protected]<mailto:[email protected]>
https://www.ietf.org/mailman/listinfo/nvo3

_______________________________________________
nvo3 mailing list
[email protected]
https://www.ietf.org/mailman/listinfo/nvo3

Reply via email to