RE: warning suggestions for draft-bberry-pppoe-credit

 

>-----Original Message-----
>From: James Carlson [mailto:[email protected]] 
>Sent: Thursday, December 08, 2005 12:10 PM
>To: Stan Ratliff (sratliff)
>Cc: [email protected]
>Subject: RE: [Pppext] warning suggestions for draft-bberry-pppoe-credit
>
>Stan Ratliff (sratliff) writes:
>> It allows the radio vendors to do what they do best, and at the same
>> time, allows router vendors to do the same. Propagation of the P2P 
>> characteristics of TDMA links back into the router allows the router
>> to allocate QoS queues on a radio link-by-radio link basis, giving 
>> good granularity and control. 
>
>The point here is that Ethernet is Ethernet, and if this link requires
>the sort of explicit flow control described here, then it's most
>certainly _not_ Ethernet.  The problem is not at this layer.
>

Agreed. The intent of the draft is to use Ethernet/PPPoE to 
allow two devices, each having its own specialty, the ability to 
communicate with each other over a ubiquitous medium (Ethernet), 
and still preserve the characteristics of the RF transmission. 


>Either the radio itself needs to be fixed to provide the service it's
>claiming to provide, or it needs to abandon the pretense.  Providing a
>service that doesn't work as expected isn't helpful.
>
>Ethernet supports multiple protocols.  What would you do if someone
>ran some protocol *other* than PPPoE over this pseudo-Ethernet?
>Wouldn't you be forced to fix the problem all over again?
>

I don't understand what you mean here. In our implementation, the 
radio and router are connected via a real Ethernet, not a pseudo-
Ethernet. There is no intersection I'm aware of with other devices 
on the Ethernet communicating with each other via different protocols. 
The intent isn't to create a pseudo-Ethernet with the extensions,
rather, 
the intent is to create a group of pseudo point-to-point links,
representing
the individual TDMA radio links, tunneling over the Ethernet connection 
from radio to router. 

>> So, at the worst, I would like to see some text in the draft that 
>> states the applicability to TDMA-like radios, as opposed to what 
>> you have here. 
>
>Please suggest some text, and I'll see what I can do with it.

Ok, here's my shot at it: 

  "The techniques described in this document will not be applicable
  to all radio implementations. This document describes extensions 
  better applied to radios with point-to-point transmission 
  characteristics (e.g. TDMA). When used with multi-access, broadcast
  capable radios such as 802.11, these techniques may have unpredictable
  interactions with higher and lower layers."

Regards,
Stan

>
>-- 
>James Carlson, KISS Network                    
><[email protected]>
>Sun Microsystems / 1 Network Drive         71.232W   Vox +1 
>781 442 2084
>MS UBUR02-212 / Burlington MA 01803-2757   42.496N   Fax +1 
>781 442 1677
>

_______________________________________________
Pppext mailing list
[email protected]
https://www1.ietf.org/mailman/listinfo/pppext