North American summit: IPv6 implementation still facing roadblocks

Although ISPs and enterprises continue to roll out IPv6, the protocol has a long way to go before it's fully implemented.

DENVER-- When it comes to IPv6, this much is clear: Fourteen years after the protocol was first deployed, progress...

has been made. But proponents readily concede they have a long way to go before the standard replaces IPv4 as the method of choice within ISPs and enterprise networks.

Some wonder why they should have to re-engineer their network just because there is something wrong with the Internet.

Patrick Hubbard,
senior technical marketing manager, SolarWinds

Last week's 2013 North American IPv6 Summit found optimism tempered by pragmatism as speakers discussed IPv6's evolution and the challenges that inhibit full-scale IPv6 implementation.

The biggest challenge is the industry itself, according to IPv6 Forum President Latif Ladid. "The enemy [that stands in the way of deployment] is the Internet community itself," he said. "We still have to convince the right people to make it happen."

Imperative for IPv6 to take root

That said, IPv6 backers believe the flood of mobile devices and the advent of Internet-aware gadgets ranging from automobiles to medical sensors make it imperative for the protocol to take root. IPv4 is quickly running out of addresses-- the remaining U.S. blocks are expected to be exhausted in the next 18 months -- and IPv6's 128-bit addressing scheme, in combination with improvements in security and performance, make it a compelling preference, advocates say.

"The Internet is upon us," Vint Cerf, Google Inc.'s chief Internet evangelist, said in a video presented at the Summit. "Mobile and set-top boxes -- the other proliferating [challenge] is automobiles, instrumentation, medical devices -- all of this will flow into the Internet; it's a large and challenging scope," he said. Some vendors are projecting as many as 50 billion devices will be connected to the Internet by 2025, a sum that simply can't be managed by the IPv4 protocol or by methods used to enable address sharing.

Deployment is only half the battle, however. Nephos6 CEO and President Ciprian Popoviciu said enterprise and ISP executives also have to be able to measure IPv6's effectiveness in order to gauge ROI. "If you're serious about IPv6, you can't just deploy it without seeing whether it's effective. You need to talk about deployment and efficiency. The key question is not whether or not to enable IPv6. It is, 'We enabled it. Then what happened?'"

Overcoming objections to IPv6 implementation

The ability to gather verifiable and repeatable results from their IPv6 migration is just one tool engineered to help enterprises and ISPs overcome their objections about adopting the protocol.

Another is the rollout of Dynamic Host Configuration Protocol Version 6 (DHCPv6) fingerprinting, which is used to identify the devices running on an IPv6 network.

"It won't solve all the problems" with managing multiple devices on a network, "but it will help," said Tom Coffeen, chief IPv6 evangelist at InfoBlox Inc. The vendor is working with the University of New Hampshire Interoperability Lab to develop a DHCPv6 database that enterprises can use to exploit the value of the addressing scheme. "We need to manage the wireless devices that are being developed and need to be addressed," he said, adding that the protocol provides much greater detail about devices without requiring any additional overhead. It also supports the generation of more detailed reports that permit the development of more efficient policies, he said. These are all critical components as bring your own device, or BYOD, gains additional traction.

See our past coverage of the IPv6 Summit

Rocky Mountain IPv6 Summit fosters IPv6 training and other conferences

2011 IPv6 Summit from the Rocky Mountain IPv6 Task Force

Coffeen said he hopes the catalog, which will be akin to the Fingerbank database that supports IPv4 standards used to identify network devices, will be operational later this year.

An informal survey SolarWinds conducted with its customers regarding their IPv6 implementation intentions, for example, found the vast majority of responders have no immediate plans to proceed, said Patrick Hubbard, SolarWinds' senior technical marketing manager.

"The benefits are not totally understood," Hubbard said. "The findings indicate that there is a disconnect between IT and top management. Some wonder why they should have to re-engineer their network just because there is something wrong with the Internet."

But top management has to be convinced to move forward, Hubbard said. Sunny Connection Networks consultant Silvia Hagen echoed Hubbard's contentions, saying that the protocol's advantages outweigh reservations.

"IPv6 has benefits that go beyond 128-bit addressing," Hagen said. "The new address architecture offers opportunities we never had with IPv4. Use IPv6 as an opportunity to clean up your network, and if you are launching new services, go with IPv6 only where ever you can.

"IPv6 is not easy; if you want to use the opportunities and minimize investment by using natural lifecycles, implementation will take at least three years, so planning is essential. But it offers lots of improvements."

 

 

 

Dig Deeper on IP Networking

PRO+

Content

Find more PRO+ content and other member only offers, here.

Join the conversation

25 comments

Send me notifications when other members comment.

By submitting you agree to receive email from TechTarget and its partners. If you reside outside of the United States, you consent to having your personal data transferred to and processed in the United States. Privacy

Please create a username to comment.

Is your company actively moving to an IPv6 implementation?
Cancel
public research institution.
Cancel
"...the protocol provides much greater detail about devices without requiring any additional overhead."

One clarification: both DHCP in IPv4 and DHCPv6 offer the opportunity to fingerprint hosts passively with no or little additional transactional overhead. This advantage is not unique to DHCPv6 and it provides no more or less detail about the host or device than DHCP in IPv4. The goal is to make sure that if such a fingerprint exists for an IPv4 DHCP host, a similar one exists for the same host that happens to be using DHCPv6 to connect to the network. In a dual-stack environment either fingerprint will suffice for BYOD reporting and policy but DHCPv6 fingerprints will become critical to maintain such reporting and policy as IPv4 eventually begins to be deprecated.
Cancel
Sorry. Should have signed the comment below. :)
-Tom Coffeen
Cancel
Surewest is behind the curve in implimenting IPV6.
Cancel
We have some issues about having to use DHCPv6 for many platforms that require some dhcp options like PBX and DVR.
Cancel
Slowly
Cancel
To me the biggest obstacle seems to be the sheer complexity of the addresses themselves.

With IP 4 you have an address that is always of the format x.x.x.x and you immediately know when you look at it whether it is a public or private address. You can easily remember the info.

When you look at an IP6 address frankly it looks a mess.
Cancel
Security is a big concern because no companies know enough.
Cancel
ISP does not have a plan or schedule for implimentation.
Cancel
Very small Co. not yet ready
Cancel
no need at this time within the enterprise
Cancel
IPv6
Cancel
Not even on our radar yet...
Cancel
All companies have to do it sooner than later. As a software company, we had to do it sooner to test our own applications' support for IPv6.
Cancel
Internal deployment complete, home deployment complete, client deployments progressing ... #GoodTimes!
Cancel
We can't. Our web hosting is 90% or more based on cPanel, and they seem uninterested in IPv6. But once they get with the program, over 8 million domains will be switched on.
Cancel
Learning about IPV6 is required
Cancel
It is not easy to implement .
Cancel
Firstly, IPv4 addresses are very easy to remember, while IPv6 numbering is very user unfriendly. It puts people off (it does me), and means potential troubleshooting could become far harder. I remember once reading why didn't they just extend IPv4 addressing by another 4 octets, which would effectively give you 64-bit addressing, which still gives us way more address space than we'll ever need.

Second, all the talk of 50bn devices on the net by 2025. The VAST majority of these will be NAT'd with a single IP. Almost all home users do it this way, and a lot of businesses. If/when we move to IPv6 I can see us presenting a single external IPv6 address, and tunneling IPv4 within that. IPv4 will be around for a very, very long time to come.
Cancel
It is good to do have a checklist or inventory on services that could support both IPv4 and IPv6 as well as services that only run on IPv6 to have a fair idea on Return on Investment(ROI) on IP Platform of various networking devices.
Cancel
Is good to have a checklist of inventory of services that run on both IPv4 and IPv6 as well as services that run only on IPv6 to enable one to have a fair idea on return on investment (ROI) on IP platform with various networking devices
Cancel
YES. the company that i am currently working is getting ready for IPv6 implementation
Cancel
actively but carefuly
Cancel
.
Cancel

-ADS BY GOOGLE

SearchSDN

SearchEnterpriseWAN

SearchUnifiedCommunications

SearchMobileComputing

SearchDataCenter

SearchITChannel

Close