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

Re: [10GBASE-T] PAR and 5 critters



Title:
George,

What I hear you saying is that even though you claim that receiver based ANEXT cancellation is possible, it is not required for the objectives. Are you saying that we must use installation or retrofit based ANEXT mitigation in order to meet the objectives?

Sterling

George Zimmerman wrote:
Sreen -
I'm glad to hear that you implicitly agree that 10GBT is feasible on
Cat-7 and on shorter distances of Cat5e & 6.  That should form the basis
from which we can go forward with a PAR & 5 criteria.

The main difference that we have had is in regards to distances on 5e &
6 relate to ability to mitigate alien NEXT, which you accounted for zero
in your analysis.  We appear to be at an impasse on this point.
Multiple vendors have presented installation-based techniques, and you
still account for zero alien NEXT mitigation.  SolarFlare added to that
receiver based techniques, which, you still account for zero.  However,
all of this is moot, since we are not asking for standardization of
these receiver-based techniques at this time.

I suggest we move on on the points we agree on.

George Zimmerman
gzimmerman@solarflare.com
tel: (949) 581-6830 ext. 2500
cell: (310) 920-3860

  
-----Original Message-----
From: Sreen Raghavan [mailto:sreen-raghavan@vativ.com]
Sent: Wednesday, July 30, 2003 4:18 PM
To: George Zimmerman; '[unknown]'
Subject: RE: [10GBASE-T] PAR and 5 critters

George:

We at Vativ always maintained that 10Gbps is theoretically feasible
    
over a
  
distance of 100 meters on CAT-7 cabling. In fact, we were the first
company
to say that. I refer you to the following URL:


    
http://grouper.ieee.org/groups/802/3/10GBT/public/mar03/sallaway_1_0303.
pd
  
f

In the same presentation, we said that the goal of 100 meter
    
transmission
  
at
10Gbps data rate is unachievable on CAT-5e and CAT-6 cables. We
    
followed
  
up
these results with a 3-PHY vendor joint presentation in May 2003. We
    
also
  
pointed out some of the practical considerations that need to be
    
addressed
  
in both of these presentations.

My disagreements with you stem from your claims regarding 10Gbps
transmission at 100 meters over CAT-5e and CAT-6 cables. You insisted
    
that
  
10Gbps data rate was achievable on 100 meter CAT-5 cables (November
    
2002
  
Plenary). I am attaching URLs to two of your presentations to
    
underscore
  
these claims.

Presentation 1:

    
http://grouper.ieee.org/groups/802/3/10GBT/public/jan03/jones_2_0103.pdf
  
Presentation 2:

    
http://grouper.ieee.org/groups/802/3/10GBT/public/nov02/diminico_1_1102.
pd
  
f

In both of these presentations, yourself & your colleagues at
    
Solarflare
  
claimed that 100 meter distance was achievable on CAT-5 cables. In
Presentation 2, page 9, Solarflare specifically claimed that 100 meter
distance was achievable on CAT-5, and that this was possible due to
    
MIMO &
  
FEXT mitigation. Of course, no presentation to my knowledge had
independently verified Solarflare's claims in these presentations with
technically repeatable results. Once again, if such presentations
    
exist,
  
please inform us.

But now you are saying that 10Gbps is doable on some cables at some
distance
in some installations. Such a vague goal cannot be the basis for
standardization work.

Sreen Raghavan
Vativ Technologies

-----Original Message-----
From: owner-stds-802-3-10gbt@majordomo.ieee.org
[mailto:owner-stds-802-3-10gbt@majordomo.ieee.org] On Behalf Of George
Zimmerman
Sent: Wednesday, July 30, 2003 1:34 PM
To: [unknown]
Subject: RE: [10GBASE-T] PAR and 5 critters


Sreen -

I will continue to treat you with respect, and assume that somehow I
misunderstand the tone of your email.  I can answer your arguments,
    
but
  
then let us focus back on the PAR & 5 criteria.

I believe that we have now discussed the shortcomings of the
    
assumptions
  
of your analysis. If you wish to go do work we can discuss it forward.

The process of building consensus is not the same as a full disclosure
of algorithms & architectures.  In the interest of moving things
forward, we have focused on building broad market potential looking at
the installed base.  The capacity results presented in the consensus
proposal assumed no alien NEXT mitigation and still addressed > 60% of
the installed base at 50meters. We can do better, but surely you can
agree that 10 Gigabit transmission is feasible on some length of Class
    
D
  
or E cabling.  Even you signed on to presentations proposing 10
    
Gigabit
  
on Class F cabling, do you now wish to withdraw that proposal?

I believe that you misunderstood me on the email - installation based
alien NEXT mitigation is found in the presentations I referenced, as I
stated.  With regards to receiver-based alien NEXT mitigation, we were
asked to present an example of this, and we presented such an example
(not a definitive tutorial, nor necessarily the best or only
    
technique)
  
at the last meeting - it appears that at least some folks on the
reflector have caught on to it.  I understand that you disagree with
    
the
  
results, but they are what they are.

With regards to complexity, system and AFE requirements are reflected
    
in
  
the Massana analysis.   I would NOT claim computational complexity
reduction is entirely due to MIMO, nor did we claim that in November,
    
as
  
we did give examples of multi-rate & efficient filter implementations.
There is a lot of literature on efficient high-rate DSP, and I believe
that others have understood this issue on the reflector, and have
disagreed with you.

Let us now focus back on the PAR & 5 criteria.  We all took a step
forward to technical feasibility with the text drafted (and voted in
    
by
  
the SG) in San Francisco.  Included within it was the feasibility on
Class F, based on statements & proposals backed by Vativ & others,
    
also
  
included were statements that Class D & E could support 10G, with
acknowledgement that specification augmentation might be in order.
These are concessions for building consensus, even without Alien NEXT
mitigation of any sort.  As far as technical feasibility is concerned,
    
I
  
believe it answers the capacity concerns raised, and I have seen
agreement & public statements by other PHY companies that this is
feasible (though not in the meeting).

As far as broad market potential is concerned, looking at something
    
like
  
99% of the installed base in 2005 being Class D & E, and between
    
60-70%
  
being less than 50meters, that represents a broad market

George Zimmerman
gzimmerman@solarflare.com
tel: (949) 581-6830 ext. 2500
cell: (310) 920-3860

    
-----Original Message-----
From: Sreen Raghavan [mailto:sreen-raghavan@vativ.com]
Sent: Wednesday, July 30, 2003 12:18 PM
To: George Zimmerman; sreen@vativ.com; 'DOVE,DANIEL J
      
(HP-Roseville,ex1)';
    
'Alan Flatman'; 'Kardontchik, Jaime'
Cc: '[unknown]'; 'Sterling Vaden'
Subject: RE: [10GBASE-T] PAR and 5 critters

George:

I have gone thru' these presentations. I do not see any of these
presentations verifying your claims on:

1. Receiver ANEXT mitigation techniques,
2. MIMO somehow collapsing the DSP complexity (a claim Solarflare
      
made
  
during Nov 2002 IEEE meeting).

You claim that above techniques are essential to the feasibility of
      
10G
    
over
CAT-5e, but you never provided any MATLAB models to the group to
      
verify.
    
In
addition, you said in November 2002 that 10G was feasible over 100
      
meters
    
of
CAT-5e. Due to careful mathematical analysis done by Vativ (and made
available to the group MATLAB models to verify) and other companies
      
over
    
past 9 months, majority in the group are now convinced that your
      
claim
  
is
    
false.

I believe that you must adhere to due scientific process to support
      
your
    
claims.

Sreen

-----Original Message-----
From: George Zimmerman [mailto:gzimmerman@solarflare.com]
Sent: Tuesday, July 29, 2003 5:52 PM
To: sreen@vativ.com; DOVE,DANIEL J (HP-Roseville,ex1); Alan Flatman;
Kardontchik, Jaime
Cc: [unknown]; Sterling Vaden
Subject: RE: [10GBASE-T] PAR and 5 critters

Presentations are given by individuals, not companies, but you can
      
find
    
them on the 10GBASE-T study group site.  Among those of note are a
January presentation from Stephen Bates, up to one just in July by
      
Shadi
    
AbuGhazaleh & Rehan Mahmood (there are more on mitigation and on
feasibility studies, Ron Nordin & Vanderlaan, Albert Vareljian,
      
Bijit
  
Halder all come to mind).

There really isn't any mystery here - going to shorter distances
increases the received signal proportionally, and mitigating alien
      
NEXT
    
decreases the noise, hence, more capacity.  What is more, because
      
the
  
crossover point (signal/noise=1) occurs at a higher frequency, only
increasing the capacity further.

George Zimmerman
gzimmerman@solarflare.com
tel: (949) 581-6830 ext. 2500
cell: (310) 920-3860

      
-----Original Message-----
From: Sreen Raghavan [mailto:sreen-raghavan@vativ.com]
Sent: Tuesday, July 29, 2003 4:24 PM
To: George Zimmerman; sreen@vativ.com; 'DOVE,DANIEL J
        
(HP-Roseville,ex1)';
      
'Alan Flatman'; 'Kardontchik, Jaime'
Cc: '[unknown]'; 'Sterling Vaden'
Subject: RE: [10GBASE-T] PAR and 5 critters

George,

Please indicate which company's presentation has independently
        
confirmed
      
your claims, and where I can find such a presentation.

-----Original Message-----
From: George Zimmerman [mailto:gzimmerman@solarflare.com]
Sent: Tuesday, July 29, 2003 3:16 PM
To: sreen@vativ.com; DOVE,DANIEL J (HP-Roseville,ex1); Alan
        
Flatman;
  
Kardontchik, Jaime
Cc: [unknown]; Sterling Vaden
Subject: RE: [10GBASE-T] PAR and 5 critters

Sreen & all -
I believe some clarification is in order.

What the presentation you reference from Portsmouth, New Hampshire
showed was that with an assumption of a high-degree of alien NEXT
        
and
    
a
      
further assumption that it could not be mitigated in any way,
        
cat5e/cat6
      
could not support 100meter operation at 10G.  This is a different
statement altogether as to whether cat5e/6 can support 10G either
        
with
    
alien NEXT mitigation, or at shorter reaches, both of which have
        
been
    
shown to yield sufficient capacity to allow 10G in numerous
presentations by multiple vendors.

The consensus proposal presented at San Francisco argued that even
without alien NEXT mitigation, there was a sufficient portion of
        
the
  
installed base of 5e & 6 coverable to merit broad market potential
        
(>60%
      
installed base at 50m or less), and that in addition to SolarFlare
showing both receiver-based (DSP) and installation-practices based
        
alien
      
NEXT mitigation examples,  other companies have now shown
        
significant
    
alien NEXT mitigation through installation practices.

These developments significantly change the capacity relations you
        
refer
      
to, making 10GBASE-T practical on the economically feasible
        
installed
    
base of cat5e & 6.

On your technical points for implementation, I respectfully
        
disagree,
    
and we have put forward our requirements, and these have been
        
confirmed
      
by at least one independent presentation.


George Zimmerman
gzimmerman@solarflare.com
tel: (949) 581-6830 ext. 2500
cell: (310) 920-3860

        
-----Original Message-----
From: Sreen Raghavan [mailto:sreen-raghavan@vativ.com]
Sent: Tuesday, July 29, 2003 1:58 PM
To: 'DOVE,DANIEL J (HP-Roseville,ex1)'; sreen@vativ.com; 'Alan
          
Flatman';
        
'Kardontchik, Jaime'
Cc: '[unknown]'; 'Sterling Vaden'
Subject: RE: [10GBASE-T] PAR and 5 critters


Dan:

We are really referring to the theory (Shannon Capacity) when we
          
say
    
10Gbps
cannot be achieved over CAT-5e or CAT-6 cabling. Theory shows
          
that
  
10Gbps
        
can be achieved over CAT-7 cabling. Practical issues to
          
accomplish
  
10Gbps
        
over CAT-7 cabling include (assuming PAM-10 modulation):

1. Building an 11-bit effective ADC at 833 MBaud,
2. Performing large number (x8 relative to 1000BaseT) of DSP
          
calculations
        
at
833MHz,
3. DDFSE critical path to be implemented in 1.2 ns
4. Building a linear transmit driver with an 833MGz bandwidth &
          
40
  
dB
      
SNR
        
The above list by no means is exhaustive, but shows the
          
implementation
      
issues that need to be considered.

Sreen

-----Original Message-----
From: DOVE,DANIEL J (HP-Roseville,ex1) [mailto:dan.dove@hp.com]
Sent: Tuesday, July 29, 2003 1:09 PM
To: 'sreen@vativ.com'; 'Alan Flatman'; 'Kardontchik, Jaime'
Cc: '[unknown]'; 'Sterling Vaden'
Subject: RE: [10GBASE-T] PAR and 5 critters

Hi Sreen,

One thing that occurs to me on this point is the difference
          
between
    
theory and application. Specifically, how many process actions
          
have
    
to
      
take place within a baud time to close the loops on the DSP and
          
what
    
process geometry would be required to make that timing closure?

I know that with 1000BASE-T, the theory was rock solid long
          
before
  
the
      
processes to implement it were reliable.

Dan
HP ProCurve

          
-----Original Message-----
From: Sreen Raghavan [mailto:sreen-raghavan@vativ.com]
Sent: Tuesday, July 29, 2003 11:52 AM
To: 'Alan Flatman'; 'Kardontchik, Jaime'
Cc: '[unknown]'; 'Sterling Vaden'
Subject: RE: [10GBASE-T] PAR and 5 critters



Just to clarify, Vativ, Broadcom & Marvell presented capacity
calculations
at the Portsmouth meeting and showed that worst-case CAT-7
(Class F) cabling
had sufficient channel capacity to achieve 10Gbps throughput
at 100 meter
distance. The reason for "may be possible" statement in the
conclusions was
that the 3 PHY vendors felt that more work needed to be done
on practical
implementation issues before the conclusion could be altered
            
to
  
a
    
more
        
definitive statement.

In addition, we proved conclusively that there was NOT
sufficient channel
capacity on existing CAT-5e (Class D), or CAT-6 (Class E)
cables to achieve
10 Gbps throughput.

Sreen Raghavan
Vativ Technologies

-----Original Message-----
From: owner-stds-802-3-10gbt@majordomo.ieee.org
[mailto:owner-stds-802-3-10gbt@majordomo.ieee.org] On Behalf
Of Alan Flatman
Sent: Tuesday, July 29, 2003 9:51 AM
To: Kardontchik, Jaime
Cc: [unknown]; Sterling Vaden
Subject: RE: [10GBASE-T] PAR and 5 critters


Message text written by "Kardontchik, Jaime"
            
Was any reason given why it would not run on Class F ? Was it
              
for
    
technical reasons or for marketing reasons ?<

The 3-PHY vendor presentation made in Portsmouth
            
(sallaway_1_0503)
    
calculated 49.36 Gbit/s capacity using unscaled Cat 7/Class F
cabling. This
figure was reduced to 37.71 Gbit/s with worst case limits.
            
Overall,
      
I
        
thought that this was a refreshingly realistic presentation
            
and
  
I
    
interpreted the summary statement "Capacity calculations with
measured data
indicate 10 Gigabit data transmission over 100m Cat 7 may be
            
possible"
        
(slide 16, bullet 3) as overly cautious engineering judgement.

So, what has changed since the May interim? Not the laws of
            
physics!
      
Best regards,

Alan Flatman
Principal Consultant
LAN Technologies