Marek,
I wish I could, but I have a conflict for the Hawaii meeting.
(Otherwise I probably would have saved my comments until then) J
I will see if I can craft a diagram prior to the meeting and if anyone finds it to be useful they will be free to add it or modify it for their presentation.
Thanks,
Hal
From: Marek Hajduczenia [mailto:marek.hajduczenia@xxxxxx]
Sent: Monday, March 05, 2012 9:05 AM
To: Hal Roberts; STDS-802-3-EPOC@xxxxxxxxxxxxxxxxx
Subject: RE: [802.3_EPOC] What is the problem with the EPoC converter, and what is the CLT?
Hi Hal,
I’d need to see a drawing to understand the second point you’re trying to make J Perhaps you could prepare a slide for upcoming meeting?
Marek
From: Hal Roberts [mailto:Hal.Roberts@xxxxxxxxx]
Sent: 05 March 2012 15:00
To: Marek Hajduczenia; STDS-802-3-EPOC@xxxxxxxxxxxxxxxxx
Subject: RE: [802.3_EPOC] What is the problem with the EPoC converter, and what is the CLT?
Marek,
“It would be nice to see at least one contribution showing a potential way of doing that at this meeting”
I agree, or several! I understand that SG is to avoid detailed technical discussions, however based on one of the “5 Criteria” the SG is also charged with showing Technical Feasibility through (preferably) multiple feasible approaches.
“…but I am sure we can come up with a way to map single-dimensional time-oriented scheduling into time and frequency domains”
I would modify this to say “…a way to map a single-dimensional time-oriented scheduling into *multiple* time and frequency domains”.
If we agree that the coax spectrum cannot support the entire EPON (much less an entire 10G-EPON), then there will be multiple “Coax Networks” (one per CNU) under one PON.
Regards,
Hal
From: Marek Hajduczenia [mailto:marek.hajduczenia@xxxxxx]
Sent: Monday, March 05, 2012 8:34 AM
To: STDS-802-3-EPOC@xxxxxxxxxxxxxxxxx
Subject: Re: [802.3_EPOC] What is the problem with the EPoC converter, and what is the CLT?
Alex,
The first question you posted got me off my feet. I think that was discussed during the CFI. MPCP is the EPON protocol we discussed about extending into coax section at the last meeting. I was not aware there were other EPON protocols that we were discussing.
What you call “unified OAM domain” puts precisely the very same set of requirements as “unified MPCP domain” – there are no changes in the way MAC control messages (be that MPCP, or OAM) would need to traverse between optical and coax domains.
Whether MPCP is suitable for scheduling CNUs or not, and how that can be achieved, I’d suggest we discuss in detail during TF meetings. This is certainly too much of a technical discussion for SG phase. It would be nice to see at least one contribution showing a potential way of doing that at this meeting, but I am sure we can come up with a way to map single-dimensional time-oriented scheduling into time and frequency domains. Again, how that could be done, I am sure we will have a chance to discuss ad nauseum during the TF meetings.
Regarding your last point. If any packet filtering is done below MAC, why would that imply changes to MAC? MAC could be blissfully unaware that anything was missing. In fact, the same mechanism is used in EPON today. ONUs filter out packets with LLIDs that are not destined for the given ONU and MAC never sees them. Yet, it is still full-duplex Annex 4A MAC, which was not modified in any way from other P2P technologies. I think there are ways of getting this function without changing MAC and MAC control layers. Again, technical details should be deferred to TF stage
Marek
From: Liu, Alex [mailto:alexliu@xxxxxxxxxxxxxxxx]
Sent: 05 March 2012 14:01
To: STDS-802-3-EPOC@xxxxxxxxxxxxxxxxx
Subject: Re: [802.3_EPOC] What is the problem with the EPoC converter, and what is the CLT?
Marek,
Can you spell out the precise benefits of extending the MPCP domain across to coax?
Is an unified OAM domain not enough to guarantee transparent provisioning between optical and coax without the need for OLT changes?
MPCP is a one dimensional (i.e., time-based) access control scheme. Mandating it on coax will rob us of the ability to account for the frequency selectivity of the RF medium, and the aforementioned “chunkiness” of spectrum availability.
Lastly, regarding the rate adaption and packet filtering functions that you mention the OCU potentially having to perform, how would you characterize these changes if not MAC changes?
Alex
<="" p="">
<="" p="">