Re: [HSSG] The List
Paul, Frank,
Given the cost argument, I think the easy way to justify
40G is to demonstrate the following.
40G performance vs 4x10LAG (accomodate TRILL,
virtualization in the analysis since we are talking
2010-2012)
40G cost vs 4x10G (assume common IC geometries,
PMD chips, optical technology, and new vs mature volumes)
The only argument against taking this approach is if you
choose to ignore the possibility that 10G will advance geometries,
consolidate multiport designs for servers, or take advantage of new
multi-channel PMD opportunities. One should also consider volume/market
maturity in their cost analysis.
When you are done, you get X% increase in performance / Y%
increase in cost.
Given my conversation with server experts in HP, I would
expect the performance number to be 1.0 - 1.5 and my experience with the silicon
and optic costs to be around .8 - 1. This means the best case is 1.5/.8 or
almost 2:1 while the worst case is 1:1. This hardly justifies a new project. I
hope someone comes in and demonstrates something with more thoroughly
developed numbers as these are ballparked.
Regarding 100G performance vs
cost, the argument supporting 100G is not performance/cost
because in the core/WAN, the ability to aggregate multiple 10G channels into a
single channel is paramount and cost is not. If you have 40Km of fiber in the
ground, digging a new trench and running 9 more fibers is not a likely
scenario and you can accomodate a reasonably high cost for that
performance level.
Dan
Paul;
I believe 100G is driven by bandwidth needs,
while 40G is primarily driven by
cost. So to demonstrate 40G
economic feasibility is paramount, but so far I did not see
good metric to justify this. I personally believe 40G
should target much more aggressive cost objective than 100G. For
example 40G may have to provide similar cost to 10G,
otherwise, there maynot exist any incentive for
end-users.
My understanding both 40g and 100g specify one same objective with
at least 100m on OM3 MMF, which is assumed based on the
installment of new fibers, this is
already relaxed compared with LRM and 10GBASE-S using various
worst-case installment fiber types. I am convinced, just like Jack
said, SFP+ 300m SR is achievable
and cost-effective than the
similar XFP case, by taking advantage of current EDC rampup
into mass production. And from EDC
vendor perspective, its true EDC is becoming a std feature of PHY IC
and its cost
premium is going to be
minimized with the std CMOS
process.
What puzzle me
is why specify multi-10G PMD lanes will
help 40G economical
feasibility. Considering various form factors, I
am with Jack, the SFP+ and SNAP12 are more viable than QSFP, also
QSFP has just mechanical specs, its
interface specs still count
on the SFP+ development. IMHO,
none of these MSAs is unique to
40G, which should be applied
equally to 100G as well, and cost
will be linearly increased with the number of channels.
Beyond PMD lanes, sound to me less
people question the 40G technical feasibility, but I still donot have clear
picture how 40G PCS layer will look like other than quad PHY,
some work similar to Mark Gustlin's CTBI. Also alternative approach for hpc using 12x3.35G
SNAP12 looks cheaper than 4x10G approach
with QSFP.
I believe we still have NOT done enough
homework in this area.
When the hssg was
initiated, there were consensus to minimize the number of PMDs to
avoid the mistake with 10G. Now we have a superPAR,
are we going to the direction eventually end up with two rates
of pretty
much everything?
Frank
Wenbin,
my presentation from
May, kolesar_01_0507, addresses my views on the intrinsic costs of items 1, 3,
and 4 of your list. Regarding item 2, I do not know how to factor in the
costs of reduced performance specs on SFP+, since that gets too subtle.
To me the issue was not if there could be cost reductions for SFP+ PMDs
but rather if SFP+ systems were capable of 10GBASE-S performance. There
have been statements that SFP+ meets 10GBASE-S specs which imply that its cost
improvements relative to XFP do not bring with them significant performance
impairments. If this is true for a broad set of suppliers and systems,
then there may be little benefit in a reduced performance 10GBASE-S spec.
Also the cost savings of a reduced performance spec would be quite
dependent upon which parameters are relaxed and would likely be quite
vendor-specific. So we may end up with little added clarity in pursuing
item 2.
In the pursuit of
clarity, I have asked for criticism of kolesar_01_0507 and for disclosure on
the impact of volume assumptions on others' relative cost claims between
4xSFP+ and QSFP. I hope that through these additional inputs a
sufficiently complete picture emerges of the costs of the available
alternatives. I see these specific issues as better communicated through
discussion on the reflector than by presentation at the meeting. But I
am open to either approach. The latter at least has the benefit of being
counted as a contribution, although I think reflector threads should also be
included in such counts.
Regards,
Paul Kolesar
CommScope Inc.
Enterprise®
Solutions
1300 East Lookout Drive
Richardson, TX 75082
Phone:
972.792.3155
Fax: 972.792.3111
eMail:
pkolesar@commscope.com
| Wenbin Jiang
<Wenbin.Jiang@JDSU.COM>
06/28/2007 01:54 PM
|
Please respond
to
Wenbin Jiang
<Wenbin.Jiang@JDSU.COM> |
|
|
To
| STDS-802-3-HSSG@listserv.ieee.org
|
|
cc
|
|
|
Subject
| Re: [HSSG] The
List |
|
There are clearly confusions/questions on the
economic feasibility of 40GE 100m objectives. Would appreciate a presentation
to compare the relative cost among the following:
1. 4x10G LAG (10GBASE-SR compliant per lane, possibly 4 x
SFP+ or QSFP, whichever costing lower?)
2. 4x10G LAG (relaxed spec from 10GBASE-SR, but similar to the 40GE per
lane basis. Possibly relaxed spec 4 x SFP+ or QSFP?)
3. 40GE compliant (presumably QSFP?)
4. 100GE compliant (SNAP12 TX+RX?)
Wenbin Jiang
From: Dove, Dan
[mailto:dan.dove@HP.COM]
Sent: Thursday, June 28, 2007 8:52
AM
To: STDS-802-3-HSSG@listserv.ieee.org
Subject: Re:
[HSSG] The List
Paul,
With regard to the point I made about a "relaxed spec for 10G" vs
writing an "entirely new Ethernet spec with multiple PMDs for 40G", it was in
response to your contention that 10GBASE-SR would be more expensive to build
than one channel of a new, unwritten, and parametrically relaxed 40G PMD. My
point was not that it is *necessary* to have a relaxed spec for 10G, but
rather that if such a relaxation is to be done, it would be less conflictive
with existing projects to do a 10G PMD than an entire set of PMDs and a 40G
MAC spec that lacks distinction from 100G in the market place. (ie: Two higher
speed Ethernets coming out at the same time with only a slight difference in
speed)
It is my firm opinion that we can build
10GBASE-SR compliant SFP+ transceivers and that the total solution cost
compared to X2 or XFP will be significantly lower in cost than today's
solution cost because of the repartitioning of the functionality, the rapidly
diminishing cost of EDC, and the port density improvements which will allow
system vendors to distribute infrastructure cost into more ports than is
possible today.
Dan
From: Paul Kolesar
[mailto:PKOLESAR@SYSTIMAX.COM]
Sent: Wednesday, June 27, 2007 4:40
PM
To: STDS-802-3-HSSG@listserv.ieee.org
Subject: Re:
[HSSG] The List
Ali,
there
seems to be a theme throughout your response that implies that cost is
paramount and compliance to 10GBASE-S is of secondary importance. This
subtle but very significant posture may be pervasive with others as well.
In my comparisons I have been holding 10G LAG implementations to the
requirements of 10GBASE-S. Dan suggested that a relaxed spec version of
10GBASE-S could be an alternative that makes LAG more attractive from a cost
perspective. And I am asking Jack to clarify his statements with respect
to compliance. Now you make similar intonations. While this may be
an alternative, it is one that distorts the reality of the day. We do
not have a relaxed spec, and although it is possible to conceive of such a
change, there is no study group activity in that direction. So it seems
mostly academic at this point because proprietary solutions of this sort do
not ensure interoperability. We need to be very clear on this assumption
in our comparisons going forward.
Regards,
Paul
Kolesar
CommScope Inc.
Enterprise® Solutions
1300 East Lookout Drive
Richardson, TX 75082
Phone: 972.792.3155
Fax:
972.792.3111
eMail: pkolesar@commscope.com
| "Ali Ghiasi"
<aghiasi@broadcom.com>
06/27/2007 12:33 PM
|
|
To
| "Jack Jewell"
<Jack.Jewell@PICOLIGHT.COM>, "Paul Kolesar"
<pkolesar@systimax.com>
|
|
cc
| STDS-802-3-HSSG@listserv.ieee.org
|
|
Subject
| Re: [HSSG] The
List |
|
Jack and Paul
The question is
not whether SFP+ can achieve 300 m SR reach similar to XFP, but how do we get
to 10G SFP+
at 2.5x the cost of 1G classic SFP for DCE (Data Center
Ethernet) with max reach of 100 m.
If we can get to 10G SFP+ at 2.5x
the cost of 1G at 300 m then the 10G PAR objective is complete, but how
long
do we wait the need is know. But I do know the combination of
lower cost optics with EDC can deliver
the 2.5x cost objective for DCE
applications near term. To get to these cost the transmitter very likely
will not be fully SR compliant and in that case it does not matter if the
reach is 100 or 300 m.
The current assumption in the HSSG is that
you can achieve SFP+ limiting performance with 4 or 10 channels without
the
use of CDR in the module, with more crosstalk, less optimum layout,
SerDes having more jitter and less tolerance compare
to small port count
PHYs, optics ??? You will get small benefit from reducing fiber reach to
100m but not enough to close the
link budget. As Dan mentioned EDC
is becoming an standard feature on PHYs and we definitely need to leverage it
for 40G/100G.
Use of linear interface is an approach that can close
the link budget without the use of CDR in the module, relax the optics
specifications, and the same interface can support passive copper Twin-ax
up to 10m.
Ali
Jack Jewell wrote:
Input from a transceiver vendor with experience and
interest in both serial and parallel modules:
First, SFP+ achieves the 300m SR objective with similar ease as
XFP. While a reduced-reach 10G PMD might reduce costs, it's a retreating
approach that requires the customer to purchase higher-cost transceivers to
fill in the reach gap, e.g. for reaches between 100m and 300m.. A
reduced operating temperature range offers immediate cost savings for some
customers without compromising reach and without any new standards
specifications. Use of a linear receiver and EDC and relaxed Tx specs
can also achieve the 300m reach at reduced cost. For this application,
the amount of compensation required is far less than what is required for LRM.
EDC is becoming widely available, to the point of being a standard
feature of PHY ICs, so the cost of this EDC is becoming insignificant.
Forward-looking standards efforts will achieve their full impact by
making use of EDC.
For QSFP,. Here
again, EDC offers great advantage, and ignoring it would be a mistake.
But crosstalk introduces module-level limitations that might motivate
reach reduction or other modification to the PMD.
Comparing costs, the SFP+ will be lower than QSFP for the
forseeable future (per lane for the same spec). The only advantage of
QSFP over SFP+ is density and possibly simpler cabling. However, a
12-channel parallel module pair, e.g. SNAP12, offers greater density than QSFP
and 2.5-3X the bandwidth at <2X the cost.
Jack
From: Dove, Dan [mailto:dan.dove@HP.COM]
Sent: Wednesday, June 27, 2007 9:32 AM
To: STDS-802-3-HSSG@listserv.ieee.org
Subject: Re: [HSSG] The List
Paul,
Regarding SFP+, I am
very familiar with the technology and have been tracking the SFF-8431
development. The architecture re-distribution of cost that SFP+ offers will
have a substantial impact on cost, especially when combined with the higher
density we can achieve with smaller geometry ASICs and multiport PHYs that
will come with it.
As for QSFP, I am less
familiar with whether or not it will provide a cost improvement over SFP+ or
be capable of meeting the existing SR spec. This is something for the QSFP
experts to consider, but like I said, a shorter 10G PMD might be the avenue to
take rather than an identity challenged 40G spec.
Regarding LAG, my conversation with HP Server architects indicates
there are a number of avenues for improvement of LAG under
development.
I cited these areas in my earlier message and
would appreciate them being addressed rather than ignored.
Dan
From: Paul Kolesar [mailto:PKOLESAR@SYSTIMAX.COM]
Sent: Wednesday, June 27, 2007 8:04 AM
To:
STDS-802-3-HSSG@listserv.ieee.org
Subject: Re: [HSSG] The List
Dan,
I can't tell how successfully SFP+ and QSFP will be
at meeting the existing 10GBASE-S spec. If they can, a new shorter
distance 10G PMD would not be of value. If they can't, then a new PMD
spec may be worth while. Those attempting to implement these lower cost
platforms need to weigh in to provide guidance. In the event that either
the QSFP and/or SFP+ can meet 10GBASE-S specs in multiple vendor's platforms,
or that a new shorter distance spec is developed that allows lower cost, the
performance issues of LAG will remain. I believe Howard's presentations
on LAG have indicated that improving LAG would not be without compromise,
leading me to conclude that, however improved, LAG performance could not
become equivalent to a 40G pipe. Developing a 40G spec would ensure a
solution that simultaneously addresses these cost and performance
issues.
Regards,
Paul Kolesar
CommScope
Inc.
Enterprise® Solutions
1300 East Lookout Drive
Richardson, TX
75082
Phone: 972.792.3155
Fax:
972.792.3111
eMail: pkolesar@commscope.com
Hi Paul,
Good points. I was not
really expecting to see a significant cost differential at the PMD although
its a good argument that a 100m PMD would be less expensive. If this is the
case, why not do another 10G PMD focused on lowering the cost of server
interconnect? I believe that would be a smaller project and have a much less
significant impact on 100G development.
Thanks,
Dan
From: Paul Kolesar [mailto:PKOLESAR@SYSTIMAX.COM]
Sent: Tuesday, June 26, 2007 6:16 PM
To:
STDS-802-3-HSSG@listserv.ieee.org
Subject: Re: [HSSG] The List
Dan,
thanks for your detailed
thoughts and proposals. I appreciate the points you made regarding the
volume effect of 10G components on the cost comparison. The presentation
I submitted for the May interim looked at the intrinsic cost factors and did
not attempt to include volume in the equation. But volume certainly can
be a significant factor. Your suggestion to look into its impact when
comparing 4x10G LAG to 40G is reasonable, but complicated at the PMD level.
As my May presentation shows there are a few ways to implement LAG on
MMF. One uses the XFP, another the SFP+, still another the QSFP.
Today the XFP is shipping to the 10GBASE-S spec, and supports 300m
transmission. Designs using SFP+ and QSFP will be more challenged to
meet this spec due to jitter, so it remains to be seen how successfully these
lower cost form factors can substitute for the XFP in 10GBASE-S compliant LAG.
However, a reduced distance requirement, such as that stated in the HSSG
objectives, would greatly improve the chances that QSFP will suffice for
"40GBASE-S". So while volume is important, these unanswered questions on
suitability make it impossible from my vantage point to determine how the
volumes for 10GBASE-S will be divided among XFP, SFP+, and QSFP. And the
effects of volume on production costs are better left to those who manufacture
the devices. Perhaps individuals with such insights will offer some
scenarios.
Regards,
Paul Kolesar
CommScope
Inc.
Enterprise® Solutions
1300 East Lookout Drive
Richardson, TX
75082
Phone: 972.792.3155
Fax:
972.792.3111
eMail: pkolesar@commscope.com
My fellow colleagues ,
Last week I sent out a list
of items that I felt need to be addressed to ensure that a 40G PAR would be
justified. At a subsequent EA teleconference intended to build concensus in
the HSSG, I offered to review the presentations made in support of 40G
Economic Feasibility and comparing 40G vs 4x10 LAG performance to ensure that
I was not being too harsh in my consideration of the material that was
presented.
Over the weekend, I reviewed every presentation I could find on
these subjects so that I could be comfortable that I was not being unfair in
my concerns. Fortunately, it was not a huge task as there are not that many to
review.
After doing so, I found myself less convinced in the
validity of some presentations that were made. This statement is not made to
criticize my colleagues, but to honor the concept of peer review which
requires that we review and criticize, otherwise we might as well just upload
them to a server and forget about them.
Specifically, I disagreed with
cost arguments made on the assumption that 10G cost remains a constant, when
in fact I anticipate substantial reductions in 10G cost over the next few
years at a rate much faster than today due to a few factors;
1)
Higher density/lower cost optical form factors (SFP+) allowing better
utilization of switch infrastructural cost and QSFP for NICs.
2) Smaller
geometry CMOS allowing higher port densities to work in synergy with PMD cost
reductions.
3) Integration of XFI / SFI interfaces directly into ASICs or
multi-port PHYs driving 10G cost further downward.
4) Higher
volumes / commoditization of 10G driving cost down much faster than the
current trajectory.
While 40G can leverage some of these elements, it
cannot leverage the volume that feeds the downward cost spiral. So in 4 years,
a 40G switch port cost is going to be based on low-volume, freshly designed
and un-amortized silicon used primarily for server interconnect, whereas a 10G
port cost will be based on amortized, high-volume silicon being used in a huge
array of applications. Having different trajectories, the relative cost for
40G will be higher than presented. This is true for 100G as well, but who is
arguing a need for 100G based on cost? It is bandwidth that drives 100G
demand.
In addition, I found presentations claiming that LAG was
insufficient to address server I/O bandwidth needs, yet those presentations
failed to address upcoming technology enhancements like TRILL and its impact
combined with I/O Virtualization, perhaps with a physical manifestation of
QSFP and MPO optics which I believe can lead to graceful performance scaling
for servers that does not demand an intermediate IEEE standard. In other
words, activities and technologies are advancing which will parse server
network access into multiple conversations that can then be put onto a LAG
group with much higher than presented performance levels.
Now, I
realize that I am swimming upstream here by asking that the proponents for
"40G now" to complete a task that took the 100G proponents almost a year
to accomplish, in less than 6 months, but then I am not asking them to do
that. My first choice, the one I proposed in Geneva, was that we move
100G forward (because it is DONE) and that we continue to work on 40G (until
it is done).
This appears to be a minority position because apparently
some people will accept an unproven 40G proposal rather than risk 100G. Others
think that 40G is proven sufficiently and are demanding "40G now" or they will
not allow a 100G PAR to go forward. Those in the latter camp must either be
unconvinced of my concerns, or they think my concerns are insufficient to
justify any further work being done to justify a 40G project.
I can
accept differences of opinion.
What I cannot do, however, is pretend
that these issues do not exist, or that the work we would have to spend
getting a 40G standard done is not going to delay the much needed 100G
aggregation solution our customers demand. I cannot ignore what I perceive as
holes in the 40G presentations.
So, to provide a little more direction
to my colleagues in the "40G now or the HSSG stalls" crowd, I am asking you to
include relative cost trajectories in your analysis of 40G vs 10G cost models,
and to include technology enhancements to LAG (TRILL, I/O Virtualization,
QSFP, MPO) in your performance analysis.
If you feel that this is
unnecessary, I am requesting that you communicate this position to me as soon
as possible so that I can prepare a presentation on these areas of concern for
the July meeting.
Respectfully,
Dan Dove
Dove
Networking Solutions - Serving ProCurve Networking by HP