draft-tequila-sls-00.txt submitted to the IETF

From: Goderis Danny (danny.goderis@alcatel.be)
Date: Tue Oct 31 2000 - 16:42:12 CET

  • Next message: Marcus Brunner: "Re: [tequila/sls] well known services"

    Hello,

    I am happy to announce that we submitted the second version of the SLS
    draft to the IETF. This (new) draft is not submitted to a particular
    working group (e.g. DiffServ WG).

    Draft Filename:
    draft-tequila-sls-00.txt

    Title:
    "Service Level Specification Semantics and Parameters"

    The draft should be available very soon.

    Summary of Changes w.r.t. the previous version
    (draft-tequila-diffserv-sls-00.txt)

    - First Notice that the abstract, introduction, section 5 (SLS
    negotiation) and 6 (security) are left UNCHANGED, although some comments
    were given on
    the mailing list, e.g. to drop the last two sections. For the time
    being, we let these topics in this draft as long as the framework
    draft is not yet available (which should capture amongst others the SLS
    negotiation stuff).

    section 3.1
    - Comments on the Scope why the semantics excludes many-to-many
    communication.
    - Some editing in the disclaimer section 3.1

    section 3.2
    - Adding : an SLS contains one and only one Flow Description (formerly
    Flow Id) (cf discussion about
    this on the mailing list)
    - set of DSCP values is also valid Flow Id/description definition
    - comments following the attribute definitions are seriously changed

    section 3.3
    - title
    - SLS allows for multi-level (n) traffic conformance (testing)
    i.s.o. only binary in/out-of-profile (allow e.g. classifying based on
    two-rate three colour marker). This also means that performance
    guarantees may be given for multiple levels (n-1) in section 3.5.

    section 3.5 :
    - adding of a time interval parameter (this needs still some
    clarifications...:-)
    - the relationship between throughput, loss and excess treatment should
    be clarified now. Throughput is an overall guarantee for all (in/out)
    packets
    - note on multi-level conformance guarantees
    - note on relationship between guarantees and conformance parameters

    section 3.8 : added is reporting guarantees, scheduled maintenance (ffs
    :-)

    Section 4
    - example 4.1 is slightly changed (delay)
    - new example 4.2: bandwidth pipe for data-services

    Regards,
    Danny



    This archive was generated by hypermail 2b29 : Tue Oct 31 2000 - 16:43:57 CET