Reader Forum: The Quiet Problem: IPv6

Editor’s Note: Welcome to our weekly Reader Forum section. In an attempt to broaden our interaction with our readers we have created this forum for those with something meaningful to say to the wireless industry. We want to keep this as open as possible, but maintain some editorial control so as to keep it free of commercials or attacks. Please send along submissions for this section to our editors at: [email protected] [email protected].
With new smartphones and other IP-enabled devices storming mobile networks around the world, it probably comes as no surprise that a majority of carriers have let a critical transition point fly below the radar for their “to do” lists. That transition is the move from IPv4 to IPv6. Despite its relatively low profile today, IPv6 will become of utmost importance to mobile operators very soon. With the demand of new devices coming online and the management of network resources, mobile operators engineering and QA departments will need to get on the IPv6 bandwagon quickly.
Many operators are unprepared for the transition to IPv6. Testing for new 3G devices alone (such as the 2 million iPads recently sold) is more than enough to keep operators up at night. With the remaining pool of available IPv4 addresses at around 7%, it’s no longer a question of if they will run out, but when. In addition, IPv4 addresses are not being disabled or un-supported – rather, they will continue to co-exist in what is known as a dual-stack environment with v6 for many years to come.
There are three major challenges mobile operators will face when they tackle the transition to IPv6.
1. Functionality migration: Mobile operators may assume that if their underlying networking infrastructure supports IPv6, the migration will execute flawlessly. This is incorrect. Many of today’s applications contain IP addresses embedded within higher-level protocols, such as SIP in Voice over IP. Without specifically developing IPv6 support for that application, it will not behave correctly when migrated. Also, the variety of operating systems that might connect to their network will all behave differently when operating in both IPv4 and IPv6 environments, not to mention the dual-stack environment. Key elements such as domain name resolution need to be thoroughly tested because all applications depend on this working flawlessly.
2. Regression/interoperability: Applications and services must be able to coexist over both v4 and v6 stacks. This means mobile carriers will need to ensure, for example, that every phone with IPv6 support is able to communicate both with other IPv6 phones of all makes and models, as well as all those supporting only IPv4. All transition mechanisms within the network will need to be exercised to ensure that each potential element in an operator’s network is going to work well.
3. Security/Resilience: Application-aware systems such as DPI, IPS and firewalls have all had years of development and patches that have gone into detecting and managing attacks and exploits over IPv4. But how do you ensure that old v4-based attacks are properly detected and handled when they come in over IPv6? And what about the tunneling schemes that would allow IPv6-based exploits to be tunneled over v4?
These security challenges will require mobile operators to have an appropriate transition plan in place. This plan will need to include adequate testing or those operators will experience unexpected and undesired consequences associated with their move to IPv6, including specific applications, services, and devices failing to work. It is not out of the realm of possibility that a complete network meltdown could occur.
What do mobile operators need to do today?
It is incumbent upon mobile operators that they re-test every element of their current networks, as well as ensure their upcoming 4G networks can support both IPv4 and IPv6 simultaneously in advance of launching them. A prudent strategy would be to implement a testing solution that allows an operator to leverage a single suite of tests which can seamlessly switch between IPv4 and IPv6 – thereby preventing the need to maintain and extend two independent sets of tests for the years to come when we will be living in a dual-stack world.
Operators may want to use this transition as an opportunity to re-evaluate their current testing strategy. There are new and innovative testing approaches that are more effective in meeting the challenges mobile operators have in testing their networks given the uniquely rapid rate of change associated with them. Relying on “canned” tests that will at best serve as an approximation of their real-world environment is a thing of the past. New and more effective approaches use the mobile operator’s own real service traffic as the basis for test creation. This not only allows tests to be created more rapidly, it also allows accurate testing of the actual services being deployed by the operator.
Unless a mobile operator adopts this new, “real-world” testing approach, they are introducing significant risk into their business by only testing an approximation of what is going to happen when their service is deployed. In today’s competitive environment, a poor product launch could cost a mobile operator hundreds of thousands of subscribers. This can easily translate to millions of dollars in lost revenue.