On 10/23/14 8:35 AM, Tom Herbert wrote:
Meta-Data: I probably missed some discussions (sorry!) but what data would
this be?
Security option to validate VNID/encapsulation headers, congestion
control data at encapsulation layer if we develop a method that uses
inband signaling, data performance optimizations (remote checksum
offload and a form of large segment offload have been proposed),
possibly a CRC to cover payload. In GUE the VNID itself is also
optional data.
Tom,
Those are useful features to consider. However, I wouldn't view them as
being part of the meta-data discussion. The meta-data discussion was
triggered by the Geneve draft which proposed ways to encode
vendor-specific data e.g. for applications like firewalls.
(I guess we need a definition of "meta-data").
Erik
_______________________________________________
nvo3 mailing list
[email protected]
https://www.ietf.org/mailman/listinfo/nvo3
