RE: Adslmib digest, Vol 1 #84 - 1 msg
Flemisch Felix <felix.flemisch <at> siemens.com>
2003-04-19 06:30:14 GMT
Hi Sabit and Ray,
thanks, Ray, for the awaited update.
From my point of view the best thing would be if you two get together immediately to replace the TBD parts in
-08 by proper contents (and inform the internet drafts editor that an update is on its way). I had some
questions on the issue but they seem not easy to be answered by e-mail.
Sabit, you announced to provide a detailed description (see attachment).
Thanks,
Felix
(on holiday until 28 April)
From: "Bob Ray" <rray <at> pesa.com>
To: <adslmib <at> ietf.org>
Date: Fri, 18 Apr 2003 09:11:29 -0500
Subject: [Adslmib] -08 revision
I've submitted the -08 revision to the internet drafts editor.
As far as I know, all that is missing are the definitions for two
objects:
vdslChanOverhead
vdshChanBurstProtection
I put "TBD" in the definitions for now. I believe Felix Flemisch
issued a request for these definitions over a month ago. I echo
this request.
If anyone needs a copy before the draft editor gets it posted,
please let me know!
Regards,
Bob Ray
PESA Switching Systems
Felix,
Sorry about not responding to this quickly.
I do not receive email from the ietf? list. Do I need to subscribe to some
email list or perhaps you can forward relevant correspondence, as you did
earlier.
I am glad to hear that not including line code specific parameters in the
basic MIB is still a valid course.
I also agree that we have missed a couple of generic parameters that specify
the configuration of the interleaver and the RS frame.
The missed parameters should be needed not only for SCM but MCM as well.
The first parameter should be called "target burst protection" and it
applies only to the Slow channel (Interleaved, if you like)
I included the qualifier "target" because the selection of linecode specific
intleaver parameters would try to achieve this target, but the result could
end up being lower or higher.
The parameter applies independently to Upstream and Downstream.
I prefer the unit microseconds, because the values are sub-ms. A reasonable
range would be 0 to 1000 microseconds in 20 microsecond increments.
The other parameter should be called "Maximum redundancy overhead" and it
applies only to the Fast channel (Not interleaved, if you like).
I included the qualifier "maximum" because the selection of linecode
specific framing parameters would use no more, but equal to or less than
this percentage, in determining the number of FEC octets.
The parameter applies independently to Upstream and Downstream.
The % unit is OK. The reasonable range would be 0 to 50% in 1% increments.
If this is agreed so far, we can provide a detailed description.
Sabit
NLC
-----Original Message-----
From: Flemisch Felix [mailto:felix.flemisch <at> siemens.com]
Sent: Sunday, March 30, 2003 2:42 PM
To: 'ssay <at> nlc.com'
Cc: 'adslmib <at> ietf.org'
Subject: dsl2002.298 and SCM LCS MIB
Hello Sabit,
Moti Morgenstern wrote on Sunday 02 March 2003 (see msg00178.html):
"For the VDSL, the DSLF decided to extract the line code specific parameters
from the basic MIB only after the members reviewed a contribution from NLC
(dsl2002.298) claiming that SCM specific parameters can be determined
according to other, generic type, parameters. The problem is that the DSLF
discovered only recently that few of those generic parameters are not
present in the MIB so configuring a SCM line is impossible without
additional parameters.
So, I suggest adding the two generic parameters Overhead (%) and Burst
Protection (ms) to the MIB, in order to allow determining the Fast Codeword
Size and FEC Size for SCM links."
But Moti is not aware of the semantics of the proposed parameters
(see msg00191.html): "NLC ... to ... provide soon the definitions
for the two alternative parameters they mentioned (i.e.,
overhead-percentage and burst protection)".
Therefore, I would like to ask you, Sabit, to resolve asap these issues
and the following questions and request of mine (see msg00183.html):
Are you proposing to add the parameters vdslChanOverhead (UNITS "%") and
vdslChanBurstProtection (UNITS "ms") to VdslChanEntry, which shall be not
for ifType interleave(124)? Please provide a formal DESCRIPTION and how
they
are used for SCM lines. After adding these two parameters, is it then clear
that basic configuration of SCM lines is possible, or might something else
be needed or useful?
If this is the case, and since these extensions are compatible with TR-057
and RFC 2662, I agree to add them to IETF's VDSL MIB and don't see any
reason
why anybody should not agree.
Best regards,
Felix Flemisch
--------------------------------------
SIEMENS AG
Information and Communication Networks
Carrier Products Systems Engineering
Hofmannstr. 51
D-81359 Munich
Tel +49 89 722 62175
e-Mail felix.flemisch <at> siemens.com
--------------------------------------
- Sabit (or his colleague) will at last find the time to provide the
detailed spec of those two generic parameters.
If nothing happens, we should return to the two "old and good" SCM specific
parameters (for which we already have detailed definitions) and move them
to the main profile table. They can and should be conditionally mandatory
for SCM applications.
Bye,
Moti Morgenstern
Senior Systems Engineer
ECI Telecoms Ltd.
Inovia Broadband Access Division
30 Hasivim St.
Petach Tikva, Israel 49133
Tel.: +972-3-9266258
Fax: +972-3-9268182
Cell: +972-55-786258
e-mail: Moti.Morgenstern <at> ecitele.com
RSS Feed