5

I currently have the idea to deploy IPv6 on the corporate network of my enterprise. But my boss (CIO) asked me what the benefits are? And I did not have an answer.

So what could I say to make the case for IPv6?

It will run in dual stack with IPv4.

We use firewalls, VPN, and multiple WAN connections.

5
  • 4
    Since one generally doesn't wake up and think "I think I'd like to deploy IPv6 today" I assume you've researched IPv6 and how its use might provide advantage to your environment. If this is so, think about what these advantages are, as well as any potential risks and if necessary translate them into terms your boss will understand. Aug 23, 2012 at 11:33
  • Well I've already deployed small IPv6 networks for myself for the following reasons: it's the future, and I like not having to NAT and having a lot of public routable IP, but here, we already have a IPv4 public network with many available IP so this argument isn't really an reason to switch.
    – Kedare
    Aug 23, 2012 at 11:45
  • 4
    The simple fact that you couldn't provide your boss with an answer means you have absolutely no reason to change your network. There is no point in making changes without a real reason. Just following a minority trend is not a valid reason. Aug 23, 2012 at 14:21
  • @John How is IPv6 a minority trend?
    – TheLQ
    Aug 24, 2012 at 18:49
  • @TheLQ, I've seen a number of reports with varying estimates of IPv6 uptake but the one thing they all have in common is that the figure is well under 10% globally. If that's not minority, what is? An out of the box support for IPv6 by an OS does not count as uptake, only full implementation does. Aug 24, 2012 at 22:39

5 Answers 5

10

A few of the benefits of IPv6, off the top of my head:

  • NAT goes away. This isn't a large issue in your organization, but it is an issue since you are likely using small bits of private address space for...
  • VPN goes away. In IPv6 you achieve the same things with IPSec, firewall rules, and...
  • Mobile IPv6. Laptop users can automatically remain connected to the corporate network with the same IPv6 address they use internally.
  • The RIR won't come back and ask you to give up some of your addresses (if they haven't approached you yet, they will soon).
  • It'll probably cost more the longer you put it off. IPv6 is not optional; everyone will have to deploy it sooner or later. "Emergency" projects can be really expensive...

Hurricane Electric has a management-level presentation The Business Case for IPv6 which you may find helpful.

It's a bit dated, but still useful: eweek's How to Build a Business Case for IPv6.

You can find much more from an Internet search.

12

If you don't know why you're making a change, or the benefits of making a change, you shouldn't be making, it period.

At the risk of making an unpopular statement, you shouldn't be promoting IPv6, let alone changing to it, because for you, you're creating a solution in need of a problem, and the only thing you'll get out of doing that is a headache. Seriously, your asking this question made me think of Joel's article on :Cats.

If you invent something that doesn't solve a problem, it better be entertaining.

And I don't know about you, but building a network from the IP scheme up does not strike me as good entertainment, particularly with IPv6 support leaving so much to be desired at present.

What you should be doing is reading up on IPv6, learning what problems it actually solves, and thinking about how implementing it can benefit your environment. Once you've done that and tested it a lab environment (so you know what new risks it creates and problems it causes), then you're ready to think about how to sell it for use in production up the chain of command.

4
  • 1
    And this is the reason it has taken 10+ years to implement IPv6
    – TheLQ
    Aug 24, 2012 at 18:49
  • 1
    @TheLQ Yeah, better idea to rush ahead blindly without knowing what you're doing and make sure IPv6 rollouts are disastrous instead. That's sure to improve adoption rates. (This site really needs an eye-rolling smiley.) The actual reason it's taken over a decade to get any level of IPv6 adoption is because it's still cheaper (and safer) to stick with IPv4, which meets the vast, vast majority of businesses needs. And business needs drive IT, not the other way around. Aug 24, 2012 at 18:55
  • @HopelessN00b The issue I see with this mentality is that absolutely nothing gets done until the ISP says "Hey, you have to use IPv6 due to shortages" or other reasons, then you would have to frantically implement IPv6 in order to have internet access. And things done frantically at the last minute usually never turn out well
    – TheLQ
    Aug 24, 2012 at 19:34
  • @TheLQ Unless you're one paying for it, it's kinda not your call, is it? Which is why business needs drive IT, and not the other way around. When you pay for environment and the upgrades and the man-hours spent switching over and debugging and supporting the "new way," you get to decide when and if that "new way" gets put in. An SA career is not about getting paid to play with the coolest new toys, it's exchanging skilled labor and knowledge that's of use to your employer for money (which is useful to you). So you do things the way they want, not because of what's "technically better." Aug 24, 2012 at 19:46
3

You can tell him that it's the future ;-)

Benefits would be, that the internal part is done, when the external part has to be done later, and so that technicians can get used to use IPv6 internally.

3
  • I already told him that is was the futur but it's not really a real advantage of reason to switch now.
    – Kedare
    Aug 23, 2012 at 11:30
  • 1
    You could also state, that it is good to do it now to make sure everything is ready, and if switches servers etc aren't ready, then make room for upgrades in next years budget plans, so it is all planned and nice :) Aug 23, 2012 at 11:35
  • 1
    Your boss is a lot smarter than you're giving him credit for. No advantage = no change. Being "the future" is not even a hint of an advantage. Aug 24, 2012 at 22:44
2

Benefits:

  1. Experience how IPv6 works now. You will need in a few years.
  2. Testing which of your HW works properly with IPv6. E.g. some switches act funny even though their manuals claim to 'do IPv6'.
  3. Plugging of IPv6 related security holes right now. (How many of your desktops run windows and have their default settings? Are they using the Teredo interface to set up a IPv6 tunnel and thus bypassing your firewalls?)
  4. You gain address space. No need to use private IP ranges. No need to use the dirty hack called NAT (and all its disadvantages)
1
  • Check. I will add this to my list of words I need to keep checking (other words on that list are acronyms such as EUFI/UEFI, which I get wrong 50% of the time)
    – Hennes
    Aug 24, 2012 at 10:38
-3

Ipv6 doesn't solve any intranet problems. 10.0.0.0/8 is big enough for any company.

It solves a problem for the entire internet, which is why it will never be deployed ever.

4
  • 1
    As much as I am opposed to people jumping on the IPv6 bandwagon for no real reason, I am even more opposed to your second sentence. It WILL happen. It's just not clear when that will be. For some people it's now, for others it may not be for another few decades but eventually it will happen. Aug 24, 2012 at 22:51
  • Well the advantage for intranet is to prevent overlapping when an enterprise is bought by another one and the network must be linked each other.
    – Kedare
    Aug 25, 2012 at 8:47
  • NATing isn't an intranet "problem" anymore? When did that happen? It's certainly a pain in my ass on our internal network, even if we're not in danger of running out of IP addresses. And as mentioned, your second sentence is just wrong. It's coming, whether anyone likes it or not. Aug 29, 2012 at 0:37
  • Well, its not going to solve any NATTing problems until the entire world is using it. which I'm still going to bet will be never.... Sep 5, 2012 at 14:44

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.