IPv6 Default Allocation - What size allocation are you giving out

Thu Oct 9 07:50:07 UTC 2014

On Thu, 2014-10-09 at 04:59 +0000, Peter Rocca wrote:
> To paraphrase a post on this list a while ago (my apologies for lack of reference).
> There are two kinds of waste:
>  - the first kind of waste is providing 'too many' subnets for someone;
>  - the second kind of waste is leaving the space unallocated forever.

Good point. But I maintain that "too many" is exactly the right number,
and not a waste at all :-)

There are only three amounts of any scarce resource - too little,
enough, and "I don't know". In an ideal world nobody knows how much disk
space, RAM, bandwidth or address space they have - they never run into
their limits. IPv6 has ticked the box for address space - why are so
many people intent on unticking it?

In my courses on IPv6, "wasted address space" *always* comes up. I
define waste as spending some finite resource for no benefit. With IPv6,
the resource is extremely abundant, though admittedly not infinite. And
the benefits from handing out big allocations are numerous:

- never resize an allocation
- never have to add an allocation
- never have to take a phone call asking to resize an allocation
- all prefixes are the same length
- easier, faster, simpler to allocate, manage, filter, firewall,
document...

... and that's just to start with. It all translates into cheaper,
easier, less error-prone. And the benefits are reaped by both parties -
the provider and the customer.

There's a case to be made, also, that simpler is more secure, because
simpler and more homogeneous networks are easier to understand, easier
to manage, and this suffer less from human error and so on.

This is what you are buying with short prefixes. There are clear
benefits, so it's not "waste".

There's another point though, that I may have made before in this forum,
and that is that whether you have 2, 200 or 2000 nodes in a /64, you are
still using, to many decimal places, zero percent of the available
address space. The number of live nodes is barely even statistical
noise. So worrying about *addresses* in IPv6 is completely pointless.

Thinking about subnets, on the other hand, does make sense - and 256
subnets (in a /56) is not very many. It's trivially easy to dream up an
entirely plausible scenario where an ordinary household chews through
that many subnets before breakfast.

Give them a /48! Give everyone a /48. There is *enough address space*
for goodness sake. All you are doing by "saving space" is putting a
completely unnecessary brake on the future - yours and theirs. Give them
more subnets, literally, than they or you know what to do with. So many
that we can't even conceive of anyone using that many. That way subnets,
like addresses, cease to be a limitation. "How many subnets do you
have?" "I don't know - does it matter?" That's where you want to be.

Don't let your limited vision limit other people. Even if YOU can't see
the point, rest assured that some bright young thing just leaving high
school will dream up something world-changingly wonderful that needs ten
thousand subnets per household...

Regards, K.

-- 
~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
Karl Auer (kauer at biplane.com.au)
http://www.biplane.com.au/kauer
http://twitter.com/kauer389

GPG fingerprint: EC67 61E2 C2F6 EB55 884B E129 072B 0AF0 72AA 9882
Old fingerprint: B862 FB15 FE96 4961 BC62 1A40 6239 1208 9865 5F9A