IPv6 is primarily known for expanding the number of Internet addresses as the current system, IPv4, is nearly exhausted due to the proliferation of IP-enabled devices like smartphones and tablets. Many of the Internet’s biggest companies participated in World IPv6 Day and the goal of the event was to raise general awareness and quantify issues such as misconfigured gear that will create broken connections for some users of IPv6. By all accounts, World IPv6 Day was the non-event everybody hoped it would be.
However, one key issue for companies is that IPv4 and IPv6 are not compatible so users assigned IPv6 addresses by their ISPs will not be able to connect to services that only use IPv4. Likewise, businesses that can only obtain IPv6 addresses will not be able to transact with customers on IPv4-only infrastructure. Services have been developed to address these challenges that help to mask compatibility issues to users, but they also introduce new challenges in delivering quality application experiences.
With the transition to IPv6 already underway, it’s critical that organizations ensure their IPv6-accessible websites perform on par with their customers’ user experience expectations. Compuware’s early analysis of IPv6-accessible sites shows that end users commonly experience slower responses times.
Since there was not a way for organizations to compare IPv6 website performance to the currently deployed IPv4 sites, last week we introduced a free IPv6 Website Performance Comparison Test. The test provides early adopters a quick and simple way to measure the response times that a user experiences when using the two protocols and gauge the impact the new protocol has on performance.
The test is executed from a dual stack IPv4/IPv6 ready machine, using a Firefox 3.5 browser to make the requests for the IPv6 and IPv4 URLs. To use the Gomez IPv6 Website Performance Comparison Test the user submits URLs for IPv4- and IPv6-enabled websites. The test produces a waterfall chart that compares the response times of each of the sites and also shows a screen capture of the IPv6 and IPv4 pages as they are seen in an actual browser. The results display the total network time for the base pages and each object on the pages.
It’s worth noting that this test is not meant to be a diagnostic tool, but more informational, relating the network times and the differences between the 2 protocols. In my next post, I’ll discuss in more detail several performance-influencing factors and what you can do to address these factors.