Transcript PPT Version

IPv6 Benchmarking
Methodology
Ciprian Popoviciu, Ahmed Hamza,
Gunter Van de Velde, Diego
Dugatkin, Bill Kine
IETF 66, March 21th 2006
Montreal, PQ
draft-popoviciu-bmwg-ipv6benchmarking-01.txt
1
Agenda
 Document Overview
 Changes made based on WG feedback
 Going forward
draft-popoviciu-bmwg-ipv6benchmarking-01.txt
2
Document Goals:
 Address a very acute and growing need for
recommendations on evaluating network element
performance for IPv6 deployments
 A complement rather than a replacement of
RFC2544 in accordance with BMWG strategy
 Provide the additional, IPv6 specific guidelines to
IP benchmarking methodology while indicating
the aspects of RFC2544 that are IP version
independent
 Maintain the structure and spirit of RFC2544.
draft-popoviciu-bmwg-ipv6benchmarking-01.txt
3
WG feedback integrated in the current version:
 Refer to RFC1242 for terminology – no specific
terminology required for this document
 Removed Section 3 as it had minor differences with
respect to RFC2544
 Re-write of Section 4 to reference the test setup
described in RFC2544
 Re-write of Section 4.1. Refer to RFC2544
appendix for Ethernet frame sizes and added an
appendix containing the frame sizes for SONET
 Removed tunneling
draft-popoviciu-bmwg-ipv6benchmarking-01.txt
4
WG feedback integrated in the current version (Cont):
 Made an explicit request for IANA address space to
be assigned for Benchmarking
 Removed the recommendation to use IMIX
 Made recommendations for benchmarking with IPv4
only, with IPv6 only and with a combination of the
two traffic types
 Removed the strict requirement for monitoring
certain device resources however we maintained a
recommendation on such a practice.
draft-popoviciu-bmwg-ipv6benchmarking-01.txt
5
WG feedback integrated in the current version (Cont):
 Hop-by-Hop EH – An optional test set for evaluating
the impact of this EH on network element resources
will be included
 Benchmarking for each EH type individually – The
only EH that requires individual evaluation is the
Hop-by-Hop. Individual evaluation of the other EHs
would provide no additional information to the data
obtained from tests using chained EHs.
draft-popoviciu-bmwg-ipv6benchmarking-01.txt
6
Interest in the document:
 Other WGs are actively providing feedback and
recommendations
 We were approached by several organizations that
expressed interest in applying these guidelines
towards evaluating products as well as
organizations interested in implementing it in their
products.
draft-popoviciu-bmwg-ipv6benchmarking-01.txt
7
IPv6 benchmarking guidelines are extremely
important due to the increasing interest in
IPv6 deployment throughout the industry
Proposed Next Steps:


Adopt the document as a WG work item
Address future comments and suggestions
draft-popoviciu-bmwg-ipv6benchmarking-01.txt
8
draft-popoviciu-bmwg-ipv6benchmarking-01.txt
THANK YOU!
draft-popoviciu-bmwg-ipv6benchmarking-01.txt
9