Re: Roy's thread
Bill Woodruff wrote:
> Rich,
>
> Hari has / will become a PMD based on the market's assumption that the three options shown below are assumed to be passive.
Bill,
I understand that you're labeling Hari as a PMD because the 8B/10B coding is passed through the "passive" PMD, but in my mind this really complicates the definition of a PMD. In the case of a WAN PHY PMD, would you consider the Hari attaching it to the PCS/PMA to be a PMD? If not, why not? It has
the same functions in either case. Hari evolved from the Serial 10 GMII presented by Howard Frazier in Montreal. It's functionality has not changed significantly since then. It's purpose was to serve as a common interface between the PCS/PMA and PMD sublayers. It's embodiment will typically be a
PCB trace between a PHY chip and a transceiver module. I believe that it is misleading to call Hari a PMD regardless of the PMD type attached to it.
> The Sun presentation in Hawaii provides 4.5ns for media skew. We can generate media PMDs for each type which are separate, but in the application the total budget needs to fit the Hari 4.5ns skew spec. Therefore Hari needs to comprehend the application of 4channel PMDs by defining a repeater.
Hari is specified to handle media skew generated by multi-channel PMDs. This was a Hari objective.
> What is a Hari repeater? It must reduce jitter without adding too much skew. Any of the three options below will need to reduce the jitter at both ends of the link.
I believe that the jitter budget was specified in the Kauai Hari Electrical presentation. This is not to say it's complete, just that it's been considered.
> What would the key specs for a Hari repeater be? It would have a jitter transfer curve with a loop bandwidth low enough to reject the jitter components which are contributed by the pattern or the system. 100KHz, 1MHz or 10MHz? The lower the LBW, the more jitter that is rejected.
>
> How about jitter transfer? How about the SONET spec of 0.1dB? How many repeaters will be allowed?
10 GbE supports only Full-Duplex point-to-point links. This means one PMD and one Hari interface at each link end. By repeater's I'm assuming that you mean the Hari element which reduces jitter.
> Is this a legitimate topic? Is there a Hari group in the IEEE, or should the Hari reflector where we should flesh this out?
So far, the informal Hari group has been taking a protocol independent view of Hari. However, since Hari has been "exposed" to the HSSG all bets are off. The HSSG should feel free to rename, chop-up, spit-out, or simply polish Hari into a useful and cost effective interface for the 10 GbE standard.
Best regards,
Rich
--
> Regards, Bill
>
> >> Bill,
>
> >> For the case where the PMD simply repeats the Hari interface and retains
> >> the 4 channels as independent I'd rather refer to this PMD option as a
> >> multi-channel PMD. We've already heard proposals on several of those:
>
> >> 1) Parallel optics: 4 channels on 4 fibers
> >> 2) CX multi-channel: 4 channels on 4 coax's
> >> 3) WWDM: 4 channels on 1 fiber
>
> >> My point is that there's more to these PMDs than just Hari and I'd rather
> >> limit Hari to being an interface rather than an interface AND a PMD.
>
> >> Best regards,
> >> Rich
>
> >> --
>
> >> Bill Woodruff wrote:
>
> >> > Roy & others,
> >> >
> >> > If we want to do a simple serial PHY, then we need a simple parallel
> >> interface. An OIF style x16 interface is good, as would be a TBI (ten
> >> bit interface). In either case we have a parallel word with clock.
> >> While this is simple, it has two issues that the Hari interface avoids;
> >> the high pin count and the tight skew requirments.
> >> >
> >> > Just because Hari exists does not make the simple x16 or x10 interface
> >> go away. Chips and modules with these interfaces either do, or can,
> >> exist, and equipment people can easily use them.
> >> >
> >> > Meanwhile the Hari world may also want a simple Hari repeater. Such a
> >> device would be a simple 'eye opener' for a media that retains the 4
> >> channels as independent. Such a repeater opens the discussion of Hari as
> >> a PHY interface, which up till now has not been well discussed.
> >> >
> >> > Regards, Bill
> >> >
> >> > Rich,
> >> >
> >> > I'm not sure what gave you the idea that Hari was favorable to serial
> >> PHYs. I do know that
> >> > it is not favorable to anything other than the legacy fiber channel
> >> 8B10B LAN PHY. Hari is
> >> > NOT a PHY neutral device interconnect. Some LAN vendors have very huge
> >> boards and they want
> >> >
> >> > something that will support their 8B10B encoding over the distance of
> >> their board, and also
> >> > the back plane if needed. By introducing Hari as the standard for
> >> device interconnect
> >> > between the PMA and the PMD, they are specifically, and possibly
> >> knowingly, hampering the
> >> > development of the agreed on WAN PHY. There is no mechanism for rate
> >> controlling over
> >> > Hari. This is in violation of the agreed objective in this regard.
> >> Hari is an attempt by a
> >> > specific camp to control the development of 10GbE and limit development
> >> of anything other
> >> > than the 10.0 only rate PHY.
> >> >
> >> > As you will remember, I called the presenters and floor on the issue of
> >> what Hari really
> >> > was. It was admitted that it was a LAN extension interconnect by the
> >> individual that
> >> > responded. For those of us that are attempting to bring 802.3 into the
> >> 20th century by
> >> > making a truly ubiquitous MAC, we have been astounded by the brazen push
> >> of this
> >> > "interconnect". Chip makers will confirm that Hari specifically makes
> >> the WAN PHY
> >> > extremely difficult to implement. Since Hari truly is not common to all
> >> of the PHYs as
> >> > specified in the objectives, I suggest that it be withdrawn from
> >> consideration as part of
> >> > the 10GbE standard.
> >> >
> >> > Thank you,
> >> > Roy Bynum
> >> >
> >> > Bill Woodruff ph: 805 496-7181 x14
> >> > GiGA North America Inc. fax: 805 496-7507
> >> > 299 W. Hillcrest Dr., Suite 206 woodruff@giga-na.com
> >> > Thousand Oaks, CA 91360 http://www.giga.dk/
> >> >
> >> > See us at 00OFC in Baltimore, March 7-9 2000, Booth 2450
----------------------------------------------------------
Richard Taborek Sr. 1441 Walnut Dr. Campbell, CA 95008 USA
Tel: 408-330-0488 or 408-370-9233 Cell: 408-832-3957
Email: rtaborek@qedinc.com or rtaborek@earthlink.net