PCE Working Group Meeting IETF-67, November 2006, San Diego

presentation player overlay
1 / 6
About This Presentation
Transcript and Presenter's Notes

Title: PCE Working Group Meeting IETF-67, November 2006, San Diego


1
PCE Working Group MeetingIETF-67, November 2006,
San Diego
  • A set of monitoring tools for Path
    Computation
  • Element based Architecture
  • draft-vasseur-pce-monitoring-01.txt
  • JP Vasseur
    (jpv_at_cisco.com)

2
Quick overview
  • PCE Monitoring WG item,
  • The need for OAM functions has been expressed at
    several occasions,
  • Main OAM features specified by this document
  • Check of PCEs liveness along a path computation
    chain (may be reduced to a single element) where
    the set of PCEs may or may not be specified by
    the PCC,
  • Performance metric collection (e.g. processing
    times, ) related to the PCE(s) involved in the
    path computation chain for a specific or a
    general request,
  • Standalone functionality or part of a path
    computation process.

3
Quick overview
  • Monitoring functions can be activated during
    path computation request (PCReq/PCRep messages)
    or as part of a standalone request (newly defined
    PCEP messages)
  • New PCEP messages PCMonReq and PCMonRep
  • New PCEP objects defined
  • MONITORING (carried within PCMonReq/PCMonRep or
    PCReq/PCRep messages) Check, Record, General,
    Processing time, Incomplete, Monitoring-id-number,
  • PCE-ID (carried within PCMonReq/PCMonRep or
    PCReq/PCRep messages) used to record the PCE IP
    address
  • PROC-TIME (carried within PCMonRep/PCRep
    messages) (Estimated) Current/Min/Max/Average/Var
    iance processing times

4
Three use cases (example shown when the BRPC
procedure is in use)
PCE
Area 2
Area 1

ABR1

ABR3
Example 1 check of the path computation chain
liveness for TE LSP T1 (no path computation),
path computation chain unknown PCE-ID optionally
recorded
B
B


ABR3
A
ABR4
Area 0
Area 2
Area 1

ABR1

ABR3
Example 2 Path computation request performance
metric along the path computation chain. PCE-ID
optionally recorded
B
B


ABR3
A
ABR4
Area 0
5
Three use case (example shown when the BRPC
procedure is in use)
Example 3 performance metric gathering for a
general request along a specific path computation
chain (ABR1-ABR4)
Area 2
Area 1

ABR1

ABR3
B
B


ABR3
A
ABR4
Area 0
6
Conclusion
  • Comments have been received on the list, new
    items to be covered in the next revision, details
    on procedures,
  • New metrics can easily be added in future
    revisions (e.g. Timestamping, queues states, )
    without any change to the architecture
  • Is there a WG support for such functionality ?
  • WG ID ?
Write a Comment
User Comments (0)
About PowerShow.com