In the January 2010 survey we received responses from 206,741,990 sites.
The biggest change this month is that 30M hostnames at qq.com expired from the survey. Last year all qq.com blogs were made public, leading to a large number of hostnames being added to the survey in February 2009. However, in the last 6 months qq.com have from time to time made all blogs private again, and have also stopped reporting blog activity to well known syndication points, so Netcraft can no longer tell how many accounts are active.
The market share for all the other major web servers has increased. Apache gained approximately 3M hostnames compared to the December 2009 survey, bringing their total to 111.3M. In second place comes Microsoft with a 600k increase. nginx lost hostnames this month but grew substantially in active sites, gaining 2.7M.Total Sites Across All Domains August 1995 - January 2010
Market Share for Top Servers Across All Domains August 1995 - January 2010Top Developers
Developer December 2009 Percent January 2010 Percent Change Apache 108,953,838 46.59% 111,307,941 53.84% 7.25 Microsoft 49,184,244 21.03% 49,792,844 24.08% 3.05 nginx 16,249,950 6.95% 15,568,224 7.53% 0.58 14,110,280 6.03% 14,550,011 7.04% 1.00 lighttpd 840,654 0.36% 955,146 0.46% 0.10
Amidst the heavy snowfall that is covering parts of southern England, National Rail’s website at www.nationalrail.co.uk has been unable to cope with the number of visitors eager to know which train services may be delayed or cancelled.
The website was unavailable yesterday afternoon and for most of this morning, with some respite overnight. The site is currently still unavailable, ironically preventing commuters from viewing National Rail’s Service Disruptions page.
Live performance graphs can be viewed here.
National Rail is the title used by the Association of Train Operating Companies, which runs the train services in Great Britain. Its website – when available – allows visitors to plan journeys, and view real time service information and fares.
It's possible that the performance problems over the past day may be attributable to the wrong type of snow, which has bedevilled British rail travellers before.
Rank Company site OS Outage
DNS Connect First
Total 1 Swishmail FreeBSD 0:00:00 0.005 0.624 0.036 0.072 0.184 2 iWeb Technologies Linux 0:00:00 0.005 0.131 0.065 0.131 0.131 3 www.westhost.com Linux 0:00:00 0.005 0.329 0.082 0.168 0.356 4 INetU unknown 0:00:00 0.010 0.077 0.036 0.079 0.139 5 www.navisite.com Linux 0:00:00 0.010 0.919 0.039 0.517 0.614 6 www.netcetera.co.uk Windows Server 2003 0:00:00 0.010 0.150 0.093 0.193 0.405 7 Kattare Internet Services Linux 0:00:00 0.010 0.315 0.103 0.207 0.519 8 New York Internet FreeBSD 0:00:00 0.014 0.325 0.038 0.082 0.209 9 Hosting 4 Less Linux 0:00:00 0.019 0.197 0.041 0.089 0.226 10 www.hostdime.com Linux 0:00:00 0.019 0.341 0.073 0.195 0.446
This is Swishmail's second consecutive appearance at the top; last month, it shared the glory with Verio. Swishmail's website is served by nginx on FreeBSD. This New York based company specialises in business email hosting, but also offers several web hosting plans, all of which include access to a webmail client and shared calendars.
iWeb Technologies' website is hosted on a Linux server in Canada, and offers dedicated, shared and colocation hosting as well as domain registration services. The company recently announced an annual revenue increase from $15M to $27M.
WestHost also uses Linux for its main site and hosts more than 70,000 other websites. Their services include shared web hosting, dedicated servers, reseller hosting and domain name registration. WestHost was acquired by UK2 Group in 2008, but continues to offer services under its own name.
Six of the most reliable hosting company sites in December were running Linux, and two were using FreeBSD. Netcetera once again provided the only Windows presence, with its site being served by Microsoft IIS 6.0 on Windows Server 2003.
Netcraft measures and makes available the response times of fifty 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.
Further information on the measurement process and current measurements are available.