[arin-ppml] The non-deployment of IPv6

[Chris Engel]

Ted,

If we are talking about switching our network infrastructure to native IPv6, even just the cost of figuring out the cost is not insignificant. Off the top of my head, I would need to factor in:

1) Hardware, OS and Software compatability .... and we have stuff ranging anywhere from brand new in beta to more then 10 years old currently in service.

2) Procedural Changes ..... have to figure out what current operating procedures would be no longer functional under IPv6 and how much work it would take to update them.

3) Contractual & Compliance Issues.... have to figure if any of the changes to our standard operating procedures mandated by such a change would have contractual or compliance side effects...and how to go about mitigating them. This would probably involve some consultation with Legal... and thier time AIN'T cheap.

4) Training & Productivity.... have to figure out what sort of training staff that would be impacted by the change in procedures might need...and also maybe the effect on productivity as staff ramped up thier learning curves.

5) Implimentation.... the actual cost of obtaining and configuring the IPv6 network(s).

Now for all that cost/effort.....what EXACTLY am I supposed to be gaining in terms of benefits out of IPv6?

It's not hard to see why we aren't jumping at IPv6. I'm certain at some point I'll have to support some connectivity to it. I'm hoping that I can achieve that by simply layering on a service here or there over our existing infrastructure. In other words, augmenting rather then supplanting our existing infrastructure....and that would be the most cost effective way to go.

Heck, considering the costs...for ALOT of Enterprises I could even see them saying...." you know what...it's cheaper to NOT be able to connect to the 5% of the internet that is IPv6 only for the next 5 years then it is worry about incurring this."

I'm pretty sure be those won't be the only options, though .... I have a feeling that there will be ALOT of  v4 to v6 &  v6 to v4 gateway service type solutions around in a few years time....as the demand for that is simply going to be too profitable to ignore.

Chris Engel
________________________________________
From: Ted Mittelstaedt [tedm at ipinc.net]
Sent: Wednesday, December 09, 2009 3:51 PM
To: Chris Engel
Cc: arin-ppml at arin.net
Subject: Re: [arin-ppml] The non-deployment of IPv6

Chris Engel wrote:
> Well, speaking for myself and the Enterprise I manage..... I see
> absolutely no reason to do IPv6 at the present time. There is no
> benefit to it and a fairly decent cost.

Chris,

   It's pretty common for people on this list to throw a lot of
generalizations around, and cost is a favorite.  Understand that
I'm not picking on you here since a lot of other orgs are saying
the same thing, but for the sake of example would you be willing
to list the specifics of what you found would be costly for your
org to support it?  Typically what stops orgs from deploying is
specific problems, not general ones, and unless we talk specifics
the discussion turns into more of a philosophical discussion about
what's right, and nothing gets solved.

   With my org our main cost was labor, that is yours truly labor
hours.  All the equipment we use supports it, the labor part was in
creating and testing configuration changes, and I did that in between
doing my real work.  (you might argue that IS my real work, also,
heh heh)  And I still have work to do, our DNS servers are overdue
for software upgrades that I keep meaning to set aside time for, but
have not yet done.

Ted