On Wed, 2010-07-21 at 18:57 +0200, Alex Band wrote:
We've been working on an exercise for the IPv6 training course [...] how to subdivide prefixes over their network and how to write an addressing plan. Here's a PDF with the exercise (two pages A3): http://bit.ly/c7jZRJ I'm curious to hear if you think it's clear and useful.
I think it is very clear, and very useful. Areas that might be improved: 1: It does not mention technical or management policy considerations. Those need to be clear before you can start divvying up your allocation. A technical policy point, for example, might be whether you are going to stick to /64 subnets throughout, or accept /126 on point to point links. A management policy might be to always provide for some level of expansion. I think the exercise would be better if it were prefaced by a phase where students (or the class as a group) work out the policies that will apply. 2: It is a very detailed, specific example using real numbers. The exercise demands a lot of technical precision that doesn't relate to the intellectual problem at hand. Technical precision is good, but it might be better sought in a separate exercise. You might get more out of students by getting them to think about what they would do in more general terms - i.e., the sizes and groupings of subnets rather than what specific subnets they would use. This would also help emphasis the point that it's not addresses that matter in IPv6 - it's subnets. 3: "There is no significant growth". Part of any network design should be to mitigate the unexpected. So IMHO students should be encouraged to, for example, duplicate addressing structures across POPs as much as is feasible, build in reserves for expansion and contraction, and use ULA addressing for internal networking components to reduce the pain if networks are split or merged. Regards, K. -- ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ Karl Auer (kauer@biplane.com.au) +61-2-64957160 (h) http://www.biplane.com.au/~kauer/ +61-428-957160 (mob) GPG fingerprint: B386 7819 B227 2961 8301 C5A9 2EBC 754B CD97 0156 Old fingerprint: 07F3 1DF9 9D45 8BCD 7DD5 00CE 4A44 6A03 F43A 7DEF