IMIX The Controversy

1
(No Transcript)
2
IMIX The Controversy

• IPsec Device Benchmarking Terminology I-D
• http//www.ietf.org/internet-drafts/draft-ietf-bmw
  g-ipsec-term-02.txt
• Michele Bustos
• Tim Van Herck
• Merike Kaeo

3
Our Original 3 Possibilities(from the list
discussion)

• 1. Imix is a strictly defined Packetmix
• - poses problems with newly introduced links and
  also IPv6, but it will work with old data
• has the advantage of providing an
  apples-to-apples comparison
• 2. Imix is -a- mix the user reports what it is
• - Hard to compare results
• Can still use the old style IMIX distributions
• 3. Imix is a Packetsweep (after renaming the mix)
• - good chance of comparable results
• - can be adjusted to the needs of the network
• - exercises the full datapath
• - more futureproof

4
The Winner is. Number 2!

• 2. Imix is -a- mix the user reports what it is
• The user MUST report the mix used so we remain
  with apples-to-apples comparison
• - Hard to compare results
• This will be eliminated with a detailed reporting
  format in the methodology draft
• - Can still use the old style IMIX distributions
• This will enable users/vendors/labs the
  flexibility to continue using IMIX uniquely

5
Other Points from List Discussion

• an optional test method, with example mix
• the tester specifies their own sizes and exact
  pattern (to match a relevant circumstance)
• if a mix is used, all sizes must be tested in a
  fixed size stream
• Ingress MTU must equal Egress MTU
• Caution small changes in the sizes may make a
  large change in the results
• Not a Magic Number!
• One or so participants of note preferred No
  Mix

6
We will continue this discussion online.
Please email the list, with comments/suggestions
or IMIX references to Ipsec-Term_at_External.Cisco.C
om