• GFP Vs GMP

    From Kwati Gurunath@21:1/5 to Huub van Helvoort on Mon Sep 20 05:22:42 2021
    On Tuesday, 1 September 2015 at 20:39:29 UTC+5:30, Huub van Helvoort wrote:
    Hello Ramakrishna,

    You replied:
    So, GFP-F/ GFP-T can perform the encapsulation of variable/ constant
    bit rate signals. And GMP is not a necessity in either of the cases.
    Indeed.
    Is there something that GMP can do but GFP cannot? If not, can we say
    GFP can replace GMP? (I am ignoring the cost, even if it is a factor driving that choice).
    Yes, GMP can be used for any new client signal. From now on it is not necessary to standardise the mapping for any new client signal.
    This was mandatory in the past, before GMP was introduced.

    So conclusion: always use GMP because it is simpler, cheaper.
    The GMP circuit can be re-used. Only the receiver and transmitter
    circuits for each new signal will be different.

    Best regards, Huub.
    On Tuesday, August 25, 2015 at 7:27:20 PM UTC+5:30, Huub van Helvoort wrote:
    Hello Ramakrishna,

    You replied:

    Just to be clear..

    For variable bit rate client signals, the two stage process is:
    step1: GFP-F would map the variable bitrate client signals onto OPUk

    Yes, this is correct.
    Note that GMP is not needed, because GFP can fill the OPU completely.

    step2: OPUk gets mapped into ODUk using GMP. Correct?

    No, the OPU is part of the ODU, the difference between OPU and ODU is
    only the overhead bytes.

    But On the other hand, in case of constant bit rate client signals,
    is GFP-T necessary followed by GMP? GMP alone seems to be sufficient? >>
    When GFP-F or GFP-T is used the GFP will completely fill the OPU
    no GMP is required.

    Best regards, Huub.


    --
    reply to hhelvooort with 2 'o's
    ================================================================
    http://members.chello.nl/hhelvoort/
    ================================================================
    Always remember that you are unique...just like everyone else...



    --
    reply to hhelvooort with 2 'o's ================================================================ http://members.chello.nl/hhelvoort/ ================================================================
    Always remember that you are unique...just like everyone else...

    Hi All.
    GMP and GFP mapping procedure have their own advantages and disadvantages. Since GMP is completely transparent in nature, any failure like link toggling on line side for short duration like 4ms/5ms will make node to send this out transparently to far end device which is connected to Gige client, due to this small glitch/failure
    on far end device router Gige interface any protocols like BGP/OSPF session gets terminated or find another root till this path gets recovered completely in their route table. Advantage of this BMP is that small latency can be achived.
    With GFP mapping used, above mentioned issue can not be seen as in GFP mapping mapping procedure terminates complete signal and so it can hold any small duration like 4/5ms duration link toggling issues so far end device do not experience this small
    duration link toggling. And disadvantage of is that as there payload termination involved, there is a processing time involved hence high latency time when to GMP mapping signal. So both GMP and GFP mapping procedures are bad and good.

    --- SoupGate-Win32 v1.05
    * Origin: fsxNet Usenet Gateway (21:1/5)