Thanks, Curtis.
I stated: "VoIP in its many emerging and integrated forms could only be tracked with packet meters and Layer 3/4 traps."
...which is effectively what you expanded on in much greater detail.
>> there is the H.323 call setup and negotiation; that is TCP.<<
And whom do propose will be monitoring all of these packets and gateways at the enterprise levels, once VoIP replaces PBX tie lines to come up with a running summary of how many calls were placed for how many minutes?
>> Next is the actual VOIP data and that is RTP/UDP and these are 60 Byte ip datagrams, encapsulated in whatever the frame encapsulation is, ie. HDLC or Frame Relay or whatever.<<
Ditto. Packet cops can only work today in a universal manner if they are tagged in a kind of SS7-associated accounting system which is a possible means of countering my assertions, granted. But that would still only apply to the PSTN-like services, in the public voice nets using VoIP as the transport.
I don't think that the IETF is ready just yet to begin developing such a trapping mechanism, however. That would simply be adding fuel to the opposition's fire. The Accounting Rate Crowd, that is. Come to think of it, the ITU is having quite a bit to say lately...
>>The complexity arises with VPNs and the methods used for that. (ip-in-ip, ie. tunnels, or "Tag", or whatever you want to call it<<
Agreed.
>>--we have had this particular discussion before ;-) ). <<
Yes, kinda, at least once. ;-) Later, Frank |
