Is it possible for the Internet to be switched off?

It would be fair to say that it's difficult to imagine how the modern business world could function without the Internet. Think back to the last time your email wouldn't work, and how everything suddenly ground to a halt. On a bigger scale, organisations like stock exchanges would be dramatically slower to conduct business, and communicating within large companies would rely on Intranets (self-contained networks which do not connect externally), or even revert back to physical memos.

For businesses and the wider economy, a major Internet disruption would be a significant blow in the short to medium term. With the global population increasingly relying on the Internet, it is only natural that some people may want to disrupt or control it. But what would it take to turn the Internet 'off'?

Is the Internet resilient?

The robustness of the Internet depends on its location. The possibility of major disruption to the Internet in the developed world is reasonably slim. This is because the Internet infrastructure in these regions is quite advanced. In the UK, for example, there are multiple Internet Service Providers (ISPs) and a number of Internet Exchange Points (IXPs). Together, these increase the number of Internet connections to regions internationally, creating a more robust network and giving data more paths to take if routes are disturbed.

As a result of the mature and competitive telecommunications market in the UK, there are more than 50 undersea cables connecting the island to the rest of the world. This means that should anyone wish to physically attack infrastructure in the UK and cut off Internet access, they would need to simultaneously target all of those cables – a virtually impossible task.

What about going after IXPs, which act as major exchange points for Internet traffic and are critical to the consistent performance of the Internet? IXPs have a number of different levels of protection in place. These measures range from using multiple locations to build redundancy, to using locations such as former nuclear bunkers to physically protect the technology within.

Let's entertain the hypothetical for a moment. Even if such a co-ordinated attack was possible, and effective, and all UK-based servers and 90% of those links to the rest of the world were compromised, data would still find its way into and out of the UK. There are so many layers of redundancy in place to protect the Internet that it's actually not enough to cut physical cables and attack specific locations – for example, attackers would also need to disrupt satellite connections

While it wasn't a manmade attack, last year's Hurricane Sandy disaster in the US illustrates the point. Along with the awful humanitarian impact, infrastructure was also affected, from roads to electricity to to the Internet. However, traffic between Europe and North America was able to re-route over different paths, despite major Internet hubs being hit with Sandy's full force. This meant that information could still be transmitted, and hopefully that made some difference for the people trying to cope with the storm.

Where does the Internet go out?

In other parts of the world, the Internet is more susceptible to disruptions. This is especially true when the number of ISPs is limited. For example, there may only be one cable connecting a region or country to the Internet and only one state-owned ISP. Therefore anyone looking to interfere with connectivity only needs to target a single connection or very few sources.

An example from earlier this year is the Internet outages in Syria. The vulnerability of the Internet in developing regions was highlighted by the disruptions seen there. In May 2013, the RIPE NCC's RIPEstat graph of the region showed two cases of Internet outages in Syria. There are a number of explanations for why it happened. It could have been the result of a technical fault at a central point, such as a cable cut. Alternatively, it could have been the result of administrative action which physically stopped the network inside the country connecting to the wider Internet. The end effect was that there was no traffic flow between Syria and the rest of the world.

What's next?

The very nature of the world and innovation means that progress is constant, and Internet-connected devices are the next big frontier for consumers and businesses across the globe. A short time ago, being able to carry portable computers with us wherever we went was exciting technology; soon everything from our refrigerators to the glasses we wear could be connected to the Internet. It's a truly exciting time, with technology developing in leaps and bounds – and as more devices are connected, the Internet will become even more crucial to our daily lives and ever more difficult to switch off.

But all of these devices will also need IP addresses, so even if the Internet can't be switched off, we still need to make sure there are enough addresses to go around, which is where IPv6 comes in. All this innovation is useless if we can't connect, and we want to make sure we don't hinder the Internet's growth – or anyone's ability to connect to it – by ensuring the widespread deployment of IPv6 now, in time to keep up with the pace of innovation.