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

Re: [802.3_400G] Finding The Balance Point(s)



Dan,

You ask some excellent questions.

As many can probably attest to, innovation within a standards development project can be a daunting task. It can be significantly easier (and faster) to develop a standard around technologies the task force are familiar and comfortable discussing.

That being said, there are times where IEEE 802.3 has been innovative in what they specify. While some of those innovations may not have gone on to see success in the market, they have often laid the foundation for our next generation of specifications.

This is a good time to discuss these topics as they certainly have a direct bearing on the time it can take to complete a project and on the level of commitment required by the task force members. Please feel free to add me to the list.

Thanks,
Brad


On Wed, Feb 12, 2014 at 10:54 AM, Dan Dove <dan.dove@xxxxxxxxxxxxxxxxxx> wrote:
All,

One of the biggest challenges we face in developing Ethernet standards is identifying the appropriate balance point for technological innovation.

For 400G, we have a number of technology elements, from PCS to CDAUI to a number of PMDs. Each element can be satisfied with a straight-forward extension of 100G technology, or some level of innovation may be taken on to achieve a longer-term solution that will achieve lower cost/power/space points, but that innovation has a cost in time and energy required to achieve it.

The question of where the balance point is for each element in the architecture is exacerbated by the fact that market need varies from "I need it now and cost is not the issue!" to "I don't need it now, but when I do, cost, power, density will be essential!".

And then we have the complexity of IEEE project realities to add to the fun. A project has a defined time-frame for getting the specification done. Having some elements being straight-forward, while others being highly-innovative, can create a challenge to the industry. Do we wait for the straight-forward parts until the highly-innovative parts are figured out, or do we jettison the more complex efforts to another project to be done at a later time? Or do we leave the straight-forward parts to MSAs or other mechanisms and focus only on the level of innovation that is justified by a standard?

As we prepare to move into the March plenary, and the natural gap in activity that goes along with a TF approval process, I am thinking that we can use this time wisely to address these questions and try to get consensus around what level of innovation for each part of this project is appropriate.

I'm just thinking out-loud a bit here and even considering hosting a bridge call to discuss this as well. If you are interested in being on the distribution list for that call, please let me know.

Regards,
--
Dan Dove
Chief Consultant
Dove Networking Solutions for Huawei
530-906-DOVE - Mobile