Re: [EFM-OAM] Pause question
Sharam,
In my opinion, the use of 802.3x PAUSE flow control is of
questionable value. I think that we added it to the standard
for the sake of having something that we could call flow
control, but we deliberately stopped short of creating a
grandiose scheme, because that would have been a never-ending
project.
As a result, 802.3x is of very limited value. Some would
argue that it is of zero value. For this reason, there
is little to no interest in extending it. There are other,
better, techniques for flow control in the world, so there is
little motivation for trying to create the ideal flow control
mechanism at layer 2. Simply adding a per VLAN capability to
802.3x wouldn't be sufficient, and we would run the risk of
falling back into the "never-ending project" trap.
Howard
Shahram Davari wrote:
> Thanks Tony,
>
> But why not. Now that Ethernet is going to the WAN, there is a need to
> flow control per-VLAN at the WAN edge.
>
> -Shahram
>
>
>>-----Original Message-----
>>From: Tony Jeffree [mailto:tony@jeffree.co.uk]
>>Sent: Tuesday, May 06, 2003 3:57 PM
>>To: Shahram Davari
>>Cc: 'stds-802-3-efm-oam@ieee.org'
>>Subject: Re: [EFM-OAM] Pause question
>>
>>
>>At 11:28 06/05/2003 -0700, Shahram Davari wrote:
>>
>>
>>>Hi,
>>>
>>>Sorry for an off topic question. Is there any per-VLAN Pause
>>>control frame defined in IEEE?
>>>
>>No! Neither (in my opinion) is it at all likely that this
>>situation will
>>change.
>>
>>
>>Regards,
>>Tony
>>
>>
>>
>
> .
>
>