Thread Links Date Links
Thread Prev Thread Next Thread Index Date Prev Date Next Date Index

Re: [STDS-802-3-400G] 答复: [STDS-802-3-400G] New 50G/100G/200G SG - Potential Impact on 802.3bs?



Brad,

 

I find this discussion very helpful and not strange.  The IEEE rules seem pretty flexible if we can get the support of 75% of the voters.  The super majority of 75% is the hard part, so it helps to build consensus when everyone knows what is going on. 

 

I like how Mark is explaining what is going on in the study groups and how it could affect the task force.  I’ve missed  some meetings (because of other meetings like you) and I appreciate his recaps about how we might slice and dice the projects. 

 

I don’t think inclusion of 200G will be the long pole in the 400G tent.  We should be able to do derivatives of the 400G work pretty easily if we can agree on the signaling.  I predict the long pole will be standardizing the 50G and 100G lane signaling.

 

Kind regards,

Scott

 

 

From: Brad Booth [mailto:bbooth@xxxxxxxx]
Sent: Monday, December 14, 2015 11:14 AM
To: STDS-802-3-400G@xxxxxxxxxxxxxxxxx
Subject: Re: [STDS-802-3-400G]
答复: [STDS-802-3-400G] New 50G/100G/200G SG - Potential Impact on 802.3bs?

 

Mark,

 

Thank you for the clarification.

 

I'm not saying that short cutting has been done or that rules have been violated, but there are a number of blurred lines and a few buzzers going off that this seems very strange. For example, your statement, "Some clear consensus that some aspects of 200G (logic primarily, maybe AUI, maybe SMF) are all potentially incremental work items that .3bs could take on."

 

While many of the players involved are the same, this clear consensus is from a SG ad hoc on influencing an existing task force. That does seem to beyond the function of a study group as defined by our rules. The objectives for that SG haven't even been voted on, and the SG hasn't even had its first initial meeting. And yet there's "consensus" for modification of the .3bs task force?

From the 802.3 rules: The main responsibility of the TF Chair is to ensure the production, and to guide through the approval and publication process, a draft standard, recommended practice or guideline, or revision to an existing document as defined by the relevant PAR. Should the ad hoc's consensus be a driving factor in modifying the .3bs PAR?

Maybe it's just me, but even if the majority of the people involved in these efforts are the same, shouldn't we be using caution?

Doesn't anyone else find that a bit strange?

 

Thanks,
Brad

 

 

On Sun, Dec 13, 2015 at 2:31 PM, Mark Nowell (mnowell) <mnowell@xxxxxxxxx> wrote:

Copying the 50 Gb/s Ethernet and NGOATH Ethernet Study Groups reflector as this discussion originated there but it jumped over to the 400Gb/s reflector and has been on-going there. Can we please copy the SG reflector on these discussions…

 

Brad,

 

I don’t think we’ve yet done anything that is changing anything against our usual practices.

 

As we’ve discussed many times, we initiated 2 study groups in order to study the two (related) topics on an optimized single lane project for 50Gb/s Ethernet and the other study group to study the multiple lane variants of that.

 

As was discussed and presented in the first ad hoc, http://www.ieee802.org/3/50G/public/adhoc/archive/nowell_120215_50GE_NGOATH_adhoc.pdf regardless of how this work gets “handled” in Task Forces it does not mean that we miss or avoid any of the requirements to justify the work (CSD, objectives, PARs).

 

The starting point of assumption has always been that these 2 SGs form 2 Task Forces.  What has emerged from the discussions is that many are realizing that some of the potential objectives for the NGOATH are essentially incremental work for the 802.3bs TF and so some work has been done to start exploring what could or should be added to the .3bs  project in a way that is beneficial (accelerates market availability, commonality of expertise, leverage of specs) without it being detrimental (schedule impact).

 

I don’t see this as a short circuiting of the process.  We primarily have to pick objectives first. This is independent at some level to where the work will be done.  Once we’ve done that we then work the project documentation to justify where the work should be done.

 

Pete Anslow and Matt Brown did a further analysis looking at  some of the possibilities of how the potential objectives could be partitioned into Task Forces http://www.ieee802.org/3/50G/public/adhoc/archive/anslow_120915_50GE_NGOATH_adhoc.pdf. Again it showed ways we could consider moving forward. But as was discussed during the presentation, this did not advocate what objectives should be adopted.

 

Right now I see:

  • some clear consensus around the various 50G single lane objectives.
  • Some clear consensus that some aspects of 200G (logic primarily, maybe AUI, maybe SMF) are all potentially incremental work items that .3bs could take on.
  • A lot of confusion around 100G objectives as people realize the challenges of the work due to PMD choices or backwards compatibility issues
  • Some early thinking that some of the multi-lane PMDs that leverage these new 50G single land work could be done with same project as single lane work (with 100G caveat from above)

 

There is nothing that says if the SG doesn’t reach 75% on an objective that it needs to keep circling that topic even it is is within scope of the SG charter.  For example if we don’t adopt some of the potential 100G objectives, that doesn’t prevent us from moving forward if there is agreement to do so 

 

As I said, I do not believe we are in any way trying to go outside of operating rules.  I’ve no interest in doing so, and I doubt John has either – and I know that David Law will step in at any time he sees something heading in the wrong direction.  If there is some consensus forming within the .3bs Task Force that they want to add some objectives which requires change of their scope and CSD, then that would have to follow the usual operating procedures for that in line with what has been done before.  Until that is seen to be happening, my assumption is that we’re following usual Study Group procedures to decide what to do and document it appropriately. 

 

Obviously this is all happening in parallel and there is some urgency in mind since, if the .3bs TF do want to do this, the sooner it happens the less schedule impact it occurs.

 

Regards…Mark 

 

From: Brad Booth <bbooth@xxxxxxxx>
Reply-To: Brad Booth <
bbooth@xxxxxxxx>
Date: Friday, December 11, 2015 at 12:59 PM
To: "802.3 400GE" <
STDS-802-3-400G@xxxxxxxxxxxxxxxxx>
Subject: Re: [STDS-802-3-400G]
答复: [STDS-802-3-400G] New 50G/100G/200G SG - Potential Impact on 802.3bs?

 

Jonathan,

 

The 25GBASE-T merging into the 40GBASE-T project is an understandable analogy. Both 25GBASE-T and 40GBASE-T had unique CSDs and PARs; therefore, merging them in relation to the resources made sense.

 

My understanding is that the NG 100G and 200G will have its own CSD and PAR. While I can understand the need to address resource limitations (as a matter of fact, I highlighted this concern back in June/July with the 802.3 officers), what I was trying to understand is why .3bs would need to modify its PAR or CSD to accommodate 200G. To me, that sounded like we were advocating pulling something out of a study group to modify an existing project.

 

Just want to understand if we're operating within our working group and sponsor rules.

 

Thanks,
Brad

 

On Thu, Dec 10, 2015 at 6:03 PM, Jonathan King <jonathan.king@xxxxxxxxxxx> wrote:

Hi Brad,

 

I think the general idea is that the project structuring should be in service to the needs of the industry, rather than the other way round. There’s no intent to force anything on to any other project, but to make best use of limited resources by putting synergistic projects together.

 

I think there was a similar recent example when 25GBASE-T was amalgamated into the 40GBASE-T project.

 

Hope this helps

jonathan

 

 

From: Brad Booth [mailto:bbooth@xxxxxxxx]
Sent: Thursday, December 10, 2015 4:22 PM
To:
STDS-802-3-400G@xxxxxxxxxxxxxxxxx
Subject: Re: [STDS-802-3-400G]
答复: [STDS-802-3-400G] New 50G/100G/200G SG - Potential Impact on 802.3bs?

 

I have a procedural question:

 

The 802.3 working group approved the formation of one study group for Next Generation 100G and 200G Ethernet. How are we supposed to handle the CSD and PAR documentation and approvals?

 

My apologies if this seems obvious to others, but in the 18 years I've been doing this, I don't ever recall forming a study group to develop one CSD and PAR and then discussing how that project gets partitioned to other projects. We've definitely taken PARs and merged them, but this sounds like creating a PAR and then splitting it to merge only specific portions with another PAR. I'm sure I must be missing something here, and would be extremely grateful if someone could explain.

 

Thanks,

Brad

 

On Thu, Dec 10, 2015 at 12:37 PM, John D'Ambrosia <jdambrosia@xxxxxxxxx> wrote:

All,

Please see the email I just sent.  Discussions are moving away from inclusion of 100GbE in 802.3bs.  There is a lot of discussion yet to address how 100GbE will be handled.  Thus, conversation is moving away from including it in 802.3bs because of the anticipated schedule hit.

 

Regards

 

John D’Ambrosia

Chair, IEEE P802.3bs 400GbE Task Force

 

From: Jeffery Maki [mailto:jmaki@xxxxxxxxxxx]
Sent: Thursday, December 10, 2015 3:25 PM
To:
STDS-802-3-400G@xxxxxxxxxxxxxxxxx
Subject: Re: [STDS-802-3-400G]
答复: [STDS-802-3-400G] New 50G/100G/200G SG - Potential Impact on 802.3bs?

 

Vipul,

 

Another important consideration is dissipated power and size. With markets waiting for technology to be feasible in their desired form factor, the time horizon for adoption is gated by dissipated power and optics technology sizing. This needs to be captured in proving the market potential in the CSD.

 

Host slots have opportunity to live far longer than modules. What we standardize for host slots (i.e., electrical interface and FEC) should be thought through carefully to limit churn and resulting market confusion. Host slots should be as enabling as possible without driving up cost for any one solution.

 

*** *** ***

The per-lane specification for 400GBASE-DR4 will be written in P802.3bs. There will be no way to stop anyone from using a single lane for 100G applications regardless of explicit 802.3 standardization or not for 100G application. What concerns me more is that these implementers will have to make their own choice of FEC and may drive defacto standards for the host (i.e., ASIC). Thus, I think it important to set industry direction on FEC more than anything at this stage for 100G applications.

 

Jeff

 

 

From: Vipul Bhatt [mailto:vbhatt@xxxxxxxxx]
Sent: Thursday, December 10, 2015 11:03 AM
To:
STDS-802-3-400G@xxxxxxxxxxxxxxxxx
Subject: Re: [STDS-802-3-400G]
答复: [STDS-802-3-400G] New 50G/100G/200G SG - Potential Impact on 802.3bs?

 

Picking up a sub-thread here, I want to express my support for the views of Chris and Helen. We are better off approaching 100G as 2x50G than as DR4 divided by 4.

 

In yesterday's ad hoc call, we discussed work partitioning of 50G/100G/200G -- how to combine efforts, avoid schedule delay, simplify editorial challenges, etc. All well and good. However, our bigger obligation is to write specs for a successful standard. The success of a standard is gauged by how widely it is adopted in the market, not by how efficiently we manage the IEEE process. Widespread adoption happens only if per-port cost of an Ethernet solution is competitively low. In practical terms, this boils down to enabling low-cost optics. Therefore, our partitioning decision should be subservient to the goal of achieving the lowest cost of optics, not hinder it. (If we can align the two, all the better.)

 

To achieve low-cost optics, the biggest lever we have is extensive re-use of 50G (25 GBaud PAM4) optics -- Nx50G, where N can be 1, 2, and 4. The case of N=1 will provide a huge volume base that the component ecosystem can amortize cost over.

 

Why not use Mx100G where M can be 1 and 2? Wouldn't 100G (50 GBaud PAM4) optics be lower in cost because it uses only one wavelength? No, not yet, despite it being a seductive idea.

 

Since the .bm days, some colleagues (including me) have been drawn to the idea of single-wavelength 100G. Later, as I supported all three .bs baselines (FR8, LR8 as well as DR4), I had an opportunity to look more closely at both 25 GBaud PAM and 50 GBaud PAM4 product development. My view is now more balanced, from a technical as well as economic perspective.

 

50 GBaud PAM will certainly have its day in the sun. We will see that day arriving in the form of reports of improving and real measurements at leading edge conferences like OFC and ECOC.

 

But in IEEE, we should write specs based on performance we can prove. For broad market success, 50 GBaud PAM still has several hurdles to cross -- design, packaging, signal integrity, yields -- with no immediate timeline in sight. Some may claim to cross these hurdles soon, and that's great, but that's not the same thing as broad supplier base.

 

It is one thing to show a forward-looking lab result; it is quite another to have manufactured components that give us characterization data we can use for developing PMD specs with correctly allocated margins. From that perspective, we are on more solid ground with 25 GBaud PAM than with 50 GBaud PAM. Let's walk before we run.

 

The DR4 argument -- "We voted for DR4 in 400G, so let's carry the underlying risks and assumptions forward" -- sounds like unnecessarily doubling down on an avoidable risk.

 

With best regards,

Vipul

 

Vipul Bhatt
vbhatt@xxxxxxxxx
+1 (408) 461-8521 mobile

 

 

On Thu, Dec 3, 2015 at 7:25 PM, Xuyu (Helen) <helen.xuyu@xxxxxxxxxx> wrote:

I agree with chriss point. 100G single lane is one of important solutions for future research and need solid work. Including this to bs will bring an expected debate and delay.

 

发件人: Chris Cole [mailto:chris.cole@xxxxxxxxxxx]
发送时间: 2015124 4:33
收件人:STDS-802-3-400G@xxxxxxxxxxxxxxxxx
主题: Re: [STDS-802-3-400G] New 50G/100G/200G SG - Potential Impact on 802.3bs?

 

John


My position on 100G single wavelength is that it belongs in a future Study Group, will require substantial technical contributions, and will cause a substantial technical debate. This is incompatible with addition into an existing project like 802.3bs, unless we are willing to completely reset the schedule.

 

Further, given the broad and substantial research effort into 100G single wavelength by the optics industry, it is best for 802.3 to let those efforts play out, and not engage in more crystal balling. 802.3 is not the best place to first report and debate fundamental research results. The right place for this are refereed journals and technical conferences.

 

Chris

 

From: John D'Ambrosia [mailto:jdambrosia@xxxxxxxxx]
Sent: Thursday, December 03, 2015 12:19 PM
To: Chris Cole;
STDS-802-3-400G@xxxxxxxxxxxxxxxxx
Subject: RE: [STDS-802-3-400G] New 50G/100G/200G SG - Potential Impact on 802.3bs?

 

So you are saying do something different than what is agreed upon for dr4? 

 

From: Chris Cole [mailto:chris.cole@xxxxxxxxxxx]
Sent: Thursday, December 3, 2015 3:16 PM
To: John D'Ambrosia <
jdambrosia@xxxxxxxxx>; STDS-802-3-400G@xxxxxxxxxxxxxxxxx
Subject: RE: [STDS-802-3-400G] New 50G/100G/200G SG - Potential Impact on 802.3bs?

 

John,

 

100GBASE-DR implies nothing about modulation format. It simply designates a single lane 500m interface. Modulation would be selected based on technical and other merit. Compatibility with a 4x standard would be a consideration but hardly an overriding one.

 

Chris

 

From: John D'Ambrosia [mailto:jdambrosia@xxxxxxxxx]
Sent: Thursday, December 03, 2015 11:58 AM
To: Chris Cole;
STDS-802-3-400G@xxxxxxxxxxxxxxxxx
Subject: RE: [STDS-802-3-400G] New 50G/100G/200G SG - Potential Impact on 802.3bs?

 

Chris,

Yes I am – anything differing than what is being done in .3bs will add delay – but your assumption was not clear from your use of 100G-DR nomenclature. 

 

However, it seems to me that there is a hard argument to use something different for a single lane approach than what is used for a x4 lane approach. 

 

John

 

From: Chris Cole [mailto:chris.cole@xxxxxxxxxxx]
Sent: Thursday, December 3, 2015 2:53 PM
To: John D'Ambrosia <
jdambrosia@xxxxxxxxx>; STDS-802-3-400G@xxxxxxxxxxxxxxxxx
Subject: RE: [STDS-802-3-400G] New 50G/100G/200G SG - Potential Impact on 802.3bs?

 

John,


You are making the assumption that if we are to have single wavelength 100G, the right answer is 50GBaud PAM4. This is unsupported either by measurements or by research results.

 

We are once again reminded that trying to predict the future by more than one technology generation ahead is a low probability of success activity and therefore should not be done in standards bodies.


Chris

 

From: John D'Ambrosia [mailto:jdambrosia@xxxxxxxxx]
Sent: Thursday, December 03, 2015 11:47 AM
To: Chris Cole;
STDS-802-3-400G@xxxxxxxxxxxxxxxxx
Subject: RE: [STDS-802-3-400G] New 50G/100G/200G SG - Potential Impact on 802.3bs?

 

Chris,

I am not getting your point here – how are we introducing further delay?  We already have DR4 in the 400G standard.  What additional delay will there be to just have a single lane implementation?

 

John

 

From: Chris Cole [mailto:chris.cole@xxxxxxxxxxx]
Sent: Thursday, December 3, 2015 2:42 PM
To:
STDS-802-3-400G@xxxxxxxxxxxxxxxxx
Subject: Re: [STDS-802-3-400G] New 50G/100G/200G SG - Potential Impact on 802.3bs?

 

Hi Steve

 

One reason we may not want to do 100GBASE-DR in 802.3bs is to not add considerably delay to the 802.3b schedule while we debate the merit of supporting measurements. And given the research results we are seeing presented in recent technical publications and conferences, we are sure to see the modulation debate re-opened, which is an even better prescription to delay 802.3bs schedule.

 

Chris

 

From: Trowbridge, Stephen J (Steve) [mailto:steve.trowbridge@xxxxxxxxxxxxxxxxxx]
Sent: Thursday, December 03, 2015 10:39 AM
To:
STDS-802-3-400G@xxxxxxxxxxxxxxxxx
Subject: Re: [STDS-802-3-400G] New 50G/100G/200G SG - Potential Impact on 802.3bs?

 

Hi Chris and Rob,

Just to play devil’s advocate here, I think it depends on the objectives agreed in the study group.

If the only stuff we have objectives to do for 200G are the same things we are doing at 400G with fewer lanes, sure, it all folds right in to P802.3bs. Specifically:

We would specify 8-lane CCAUI-8 and 4-lane CCAUI-4 C2C and C2M interfaces.

We would specify 200GBASE-SR8, 200GBASE-DR2, 200GBASE-FR4, and 200GBASE-LR4 PMDs.

But the flies in the ointment would be if we have objectives to build a 200GBASE-SR4, 200GBASE-CR4, or 200GBASE-KR4 PMD, which I think would be quite challenging on the current P802.3bs schedule.

 

Presumably the reason you think 100G belongs with 50G is that you assume this project needs to do interfaces like 100GBASE-SR2, 100GBASE-CR2, 100GBASE-KR2. But why wouldn’t you do, for example, a 100GBASE-DR interface in the P802.3bs project?

Regards,

Steve

 

From: Rob (Robert) Stone [mailto:rob.stone@xxxxxxxxxxxx]
Sent: Thursday, December 03, 2015 11:10 AM
To:
STDS-802-3-400G@xxxxxxxxxxxxxxxxx
Subject: Re: [STDS-802-3-400G] New 50G/100G/200G SG - Potential Impact on 802.3bs?

 

Chris I agree with your observation, and I have been thinking the same thing with respect to the partitioning of the work.

 

It would seem that if possible, using the KR4 FEC for 50 and 100G would have a lot of benefits with respect to compatibility with existing MAC rates and 25G based technologies. I would expect that it is likely we will see co-existence of 25 and 50G / lane technologies within the same environment, and if so we should make an effort when defining the logic to enable straightforward low power connections between the different generations. Using end – end KP4 FEC would help facilitate that.

 

Thanks

 

Rob

 

From: Chris Cole [mailto:chris.cole@xxxxxxxxxxx]
Sent: Thursday, December 03, 2015 8:53 AM
To:
STDS-802-3-400G@xxxxxxxxxxxxxxxxx
Subject: Re: [STDS-802-3-400G] New 50G/100G/200G SG - Potential Impact on 802.3bs?

 

The idea of rolling 200G into the 400G project is compelling. In prior discussions, we had rejected this as too late for 802.3bs TF, so it’s encouraging to see we are willing to revisit. One mental test of why this makes sense is to consider what we would have done in 400G Study Group if we knew what we know now. Given the CFI support, it could be argued that most people would have supported both 200G and 400G. If anything, 200G is more compelling.


However, the inclusion of 100G in the 400G project is much less clear. Several arguments can be made that it more naturally belongs in the 50G project. An important consideration is that for both 50G and 100G, backwards compatibility with existing 25G I/O interfaces is important. Specifically we will want 50G supported with 2x25G electrical I/O and 100G to be supported with 4x25G I/O (CAUI-4). This suggests that for 50G single wavelength and 100G WDM2 solutions, KR4 is the appropriate FEC. For both, this gives reasonable optical margin with PAM4 modulation. On the other hand, for 200G WDM4 KP4 is a better choice, same as for 400G WDM8.


Adding only 200G to the 400G project makes for much cleaner documentation modification. Broad market potential for 200G is the same as for 400G. We don’t need to introduce new justification which is required for 100G. Not having 100G, removes any dependence on any other project, so there is no need to discuss this.

 

Chris

 

From: John D'Ambrosia [mailto:jdambrosia@xxxxxxxxx]
Sent: Thursday, December 03, 2015 6:00 AM
To:
STDS-802-3-400G@xxxxxxxxxxxxxxxxx
Subject: [STDS-802-3-400G] New 50G/100G/200G SG - Potential Impact on 802.3bs?

 

Dear Task Force Participants,

This email is to make sure that everyone is aware of conversations happening in the 50/100/200G Study Group Phone Conference that happened yesterday – Dec 2.  There has been discussion at how the multi-lane 100G/200G solutions might be rolled into the 802.3bs project.  To that end – I gave a presentation at the conference call that looked at potential modifications / additions to our PAR / CSD.  See http://www.ieee802.org/3/50G/public/adhoc/archive/dambrosia_120215_50GE_NGOATH_adhoc_v2.pdf

 

I encourage everyone to review this presentation and consider the findings on the last few pages.  Individuals may wish to participate in the upcoming 50/100/200G ad hoc calls that Mr. Nowell has planned.  For more information see http://www.ieee802.org/3/50G/public/adhoc/index.html.

 

I will be working on the meeting announcement for the January interim, and anticipate that there will be a joint session of our Task Force with the Study Groups to further consider these implications.

 

Regards,

 

John D’Ambrosia

Chair, IEEE P802.3bs 400GbE Task Force