Note: Outage times display the minimum outage time which may understate each outage by up to 15 minutes, which is the sampling frequency.
If you are researching prospective hosting locations, or performing competitor analysis and would like to buy bespoke performance monitoring of sites of your choice, or access to historical data, please mail us at firstname.lastname@example.org
Interpreting the Tables
Netcraft makes conventional HTTP GET requests to the URL exactly as shown, without providing a full OCSP request payload. If no response is received, or if the HTTP response has a status code corresponding to an HTTP error code, the response is considered failed.
Real-world OCSP performance may vary, with the caching behaviour of the Content Distribution Networks commonly used by OCSP responders influencing the results obtained by Netcraft.
In particular, to compare OCSP responders in the table above, consider the number of failed requests and the average time taken to establish a TCP connection. These metrics capture network-level failures and performance, and provide a view to the reliability of the networks used by OCSP responders.
By default the sites are ranked in order of failed requests and time to connect, shortest first, in order to give the clearest indication of network capacity and congestion, with the least impact from the performance of the individual web server, though it is possible to sort by other columns by clicking on the column heading.
Presently our performance collectors are located at New York/New York Internet-2, Italy/Aruba-2, Amsterdam/Webair-2, Singapore/DigitalOcean and Ireland/Amazon. Sites located near one of our collectors may have an advantage over others listed in the table, though this effect is mitigated by using averages collected from multiple points. Sites exclusively hosted in the Far East are at a disadvantage as we do not currently have a performance collector in the region.
Sites using a Content Distribution Network, such as Akamai, have a particular advantage as the response to our request would come from the node closest to each performance collector.
More information can be found in the FAQ.