Hi Thomas:

Hmmm, so what you want is the notion of an attachment point to a particular 
layer, in this case to explicitly identify the overlay as the network attached 
to.... (?, just checking)

>From the tenants POV it cannot tell the difference, it just sees its network, 
>so I'm not sure the distinction is as useful as it sounds...

Cheers
Dave

-----Original Message-----
From: [email protected] [mailto:[email protected]] On Behalf Of Thomas 
Narten
Sent: Wednesday, April 17, 2013 1:12 PM
To: David Allan I
Cc: Black, David; [email protected]; Eric Gray
Subject: Re: [nvo3] NVO3 Terminology changes

David Allan I <[email protected]> writes:

> I have some sympathy with Eric's point but would suggest inverting the 
> relationship to address your concern. The generic class is NIC, which 
> may or may not be virtualizated....the virtual form being associated 
> with VMs, and the generic class being associated with addressed end 
> systems in general.
>  
> I think moving away from NIC to produce some gratuitous new term is 
> not doing us any favors...

I think a TSI is roughly equivalent to a NIC, but from the perspective of a 
Tenant System. I view it as the entity that the Tenant System sees. To the 
tenant, it looks like a NIC, and is how the Tenant System interacts with the 
(virtual) network.

Using some other term than NIC or vNIC is (IMO) useful because this is a Tenant 
System entity, whereas a NIC/vNIC applies when network virtualization is not in 
use. It is good to have a term that distinguishes those cases.

Maybe a "Tenant NIC" (or TNIC) would make a better name than TSI. (Not sure 
it's a better term, but worth thinking about, especially if this really is 
essentially a "NIC".)

Thomas

_______________________________________________
nvo3 mailing list
[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