Qube Managed Services had the most reliable hosting company site in September 2013, with not a single failed request throughout the whole month, and an average connection time of only 0.066 seconds. Qube is based in London, but they also host services from data centers in New York and Zurich. Their
New York data center is at 111 8th Avenue, which is adjacent to a trunk
dark fiber line. This building is the city’s third largest in terms of floor area and was bought by Google for $1.9 billion in 2010. Qube provides managed and colocated hosting services from each of its data centers, as well as virtual data centers based on
VMware vCloud Director.
Including September, Qube (“Qualified By Experience”) has made five appearances within the top ten so far this year, and also attained another first place result in
May.
In second place, with just one failed request, was
Kattare Internet Services. It is among the most reliable sites monitored by Netcraft, managing 99.993% uptime over the past year and 99.97% over the past seven years. On September 21st, it was predicted that a damaging storm would hit the Pacific Northwest (Kattare’s base of operations), causing thunder and wind storms, with 50mph gusts strong enough to take down trees. The next day this storm took out Kattare’s power supply; however, the use of generators meant there were no outages recorded during the storm.
Hosting 4 Less narrowly missed out on second place, as although it only had one failed request during September, its average connection time was 3 milliseconds slower than Kattare’s.
Seven of September’s top ten hosting company sites were running on Linux operating systems. Five of these (including Qube, Kattare and Hosting 4 Less) used the Apache web server software, while ServerStack and krystal.co.uk used
nginx.
Netcraft measures and makes available the response times of around forty leading hosting providers’ sites. The performance measurements are made at fifteen minute intervals from separate points around the internet, and averages are calculated over the immediately preceding 24 hour period.
From a customer’s point of view, the percentage of failed requests is more pertinent than outages on hosting companies’ own sites, as this gives a pointer to reliability of routing, and this is why we choose to rank our table by fewest failed requests, rather than shortest periods of outage. In the event the number of failed requests are equal then sites are ranked by average connection times.
Information on the measurement process and current measurements is available.
