Benchmarking%20Methodology%20WG%20(bmwg)%2064th%20IETF%20 - PowerPoint PPT Presentation

About This Presentation
Title:

Benchmarking%20Methodology%20WG%20(bmwg)%2064th%20IETF%20

Description:

Core Router Accelerated Life Testing I-Ds (Poretsky, 15 min) Milestones Status (Chairs, 5 min) ... Apr 06 Router Accelerated Test Terminology I-D to AD Review ... – PowerPoint PPT presentation

Number of Views:56
Avg rating:3.0/5.0
Slides: 13
Provided by: Morton
Learn more at: https://www.ietf.org
Category:

less

Transcript and Presenter's Notes

Title: Benchmarking%20Methodology%20WG%20(bmwg)%2064th%20IETF%20


1
Benchmarking Methodology WG (bmwg)64th IETF
Vancouver, Canada
  • Monday, November 7, 2005, 1510-1710 (Oak)
  • Chairs
  • Kevin Dubray (kdubray_at_juniper.net)
  • Al Morton (acmorton_at_att.com)
  • If you are not subscribed to the BMWG mailing
    list and would like to be, put an asterisk () by
    your name when you sign the blue attendance
    sheet.  

2
BMWG Agenda
  • Note-Taker(s), IPR, and Blue Sheets
  • WG Status with some discussion (Chairs, 20
    min)
  • IPsec Benchmarking Terms New Methods (Kaeo, 20
    min)
  • IGP-Dataplane Convergence I-Ds. (Poretsky, 15
    min)
  • Core Router Accelerated Life Testing I-Ds
    (Poretsky, 15 min)
  • Milestones Status (Chairs, 5 min)
  • Revised Charter Discussion (Chairs, 15 min)
  • New Work (Chairs, Poretsky, 15 min)

3
BMWG Activity
  • AD/IESG Review
  • this space is blank for now
  • WG Last Call
  • ltdraft-ietf-bmwg-benchres-term-06.txtgt, WGLC
    ended 10/10
  • ltdraft-ietf-bmwg-igp-dataplane-conv-term-08.txtgt
    LC ended 10/26
  • ltdraft-ietf-bmwg-igp-dataplane-conv-meth-08.txtgt
  • ltdraft-ietf-bmwg-igp-dataplane-conv-app-08.txtgt
  • ltdraft-ietf-bmwg-dsmterm-11.txtgt, begin WGC
    shepherding
  • ltdraft-ietf-bmwg-ipsec-term-07.txtgt will proceed
    with Meth
  • I-Ds
  • ltdraft-ietf-bmwg-ipsec-meth-00.txtgt new
  • ltdraft-ietf-bmwg-hash-stuffing-04.txtgt WGLC
    requested.
  • ltdraft-ietf-bmwg-acc-bench-term-07.txtgt Ready for
    WGLC?
  • ltdraft-ietf-bmwg-acc-bench-meth-04.txtgt Ready for
    WGLC?

4
BMWG Activity
  • I-Ds
  • ltdraft-ietf-bmwg-acc-bench-meth-ebgp-00.txtgt,
  • ltdraft-ietf-bmwg-acc-bench-meth-opsec-00.txtgt,
  • ltdraft-ietf-bmwg-bgpcmeth-00.txtgt expected soon?

5
BMWG Activity
  • Expired BMWG I-Ds
  • ltdraft-ietf-bmwg-fib-meth-03.txtgt, No Prog
    Dead
  • ltdraft-ietf-bmwg-benchres-method-00.txtgt Pending
    term
  • New Work proposals.
  • LDP Convergence expired, but
  • draft-poretsky-protection-term-00.txt (adopt?)
  • draft-poretsky-mpls-protection-meth-04.txt
    (adopt?)
  • RFC Editor none.
  • Charter Update
  • Need more discussion.
  • New RFC

6
Revised Milestones
  • Fmr Past Due Milestones
  • Dec 05 IPsec Device Benchmarking Terminology to
    AD Review
  • Apr 06 Net Traffic Control Benchmarking
    Terminology to AD Review
  • Apr 06 Methodology for FIB related Router Perf.
    Bench. to AD review.
  • Apr 06 IGP/Data-Plane Terminology I-D to AD
    Review
  • Apr 06 IGP/Data-Plane Methodology and
    Considerations I-Ds to AD Review
  • Apr 06 Router Accelerated Test Terminology I-D
    to AD Review
  • Jul 06 Basic BGP Convergence Benchmarking
    Methodology to AD Review.
  • On Track, but running late
  • Feb 06 Router Accelerated Test Method. and
    Consid. I-Ds to AD Review
  • Feb 06 Hash and Stuffing I-D to AD Review
  • Dec 05 Net Traffic Control Benchmarking
    Methodology to AD Review (no change)
  • New
  • Apr 06 Router Accelerated Test Method for EBGP
    to AD Review
  • Apr 06 Router Accelerated Test Method for
    Operational Security to AD Review
  • Dec 05 IPsec Device Benchmarking Methodology to
    AD Review

7
Charter Text (Goal/Scope)
  • Current goal is very broad
  • to make a series of recommendations concerning
    the measurement of the performance
    characteristics of various internetworking
    technologies
  • further, these recommendations may focus on the
    systems or services that are built from these
    technologies.

8
Charter Text (continued)
  • Should we tighten this coverage?
  • to make a series of recommendations concerning
    the measurement of the performance
    characteristics of various internetworking
    technologies, primarily those developed by the
    IETF
  • Could also say preferably or exclusively or
    ??

9
What BMWG benchmarks are not, or cannot be
  • (much discussion last time)
  • Conformance tests
  • IETF consensus avoids legal entanglements (?)
  • Benchmarking methodology assumes compliance, and
    the procedures should make it difficult to cheat.
  • Reliability tests?
  • How well do these tests fall under our benchmark
    umbrella?
  • Acceptance tests
  • Implication of success criteria must be avoided
  • Operational Network tests (not controversial)
  • Division of Labor between other organizations.

10
Backup
11
Charter Text
  • Central Tenet
  • Notion of a Benchmark
  • BMWG Benchmark
  • A prescribed measurement thats a basis for
    performance evaluation AND comparison.

12
Benchmark Required Attributes?
  • Relevant and meaningful.
  • Unambiguous specification of an attribute (or
    attributes) being examined, its collection and
    reporting.
  • Universal applicability within class. (No bias
    toward or against a specific implementation.)
  • Repeatability.
Write a Comment
User Comments (0)
About PowerShow.com