Web Hosting News

Please rate this article

(No Ratings Yet)

 Loading ...

When web hosting first started some years ago, customers were not picky about the quality of the service over time. Instead they just wanted to find a web host who is affordable for hosting their web site (mostly static). When web hosting started developing as a business with high margins, there were virtually thousands of web hosts who offered the services for a cheap price. Then the factors for distinguishing the best service providers started to change. First it was just the price and then it evolved to a level where the customers were looking for characteristics such as reliability, availability, how fast the queries are answered, the flexibility of services offered etc.

Availability, which is defined for web hosting as the implementations associated for ensuring a certain absolute operational continuity during a period of measurement. This is the ability of the client to access the system. The time that the web service is not available is called downtime. There are two main downtimes as planned and unplanned. Planned downtime results due to system maintenance and it cannot be avoided with the current system design. Planned downtime makes the operational continuity disrupt, but it is communicated in advance to all the effected parties. Planned downtimes may include installation of software patched that requires rebooting of the server, upgrade of software versions (including supporting software for application software), changing or upgrading hardware components etc.

Unplanned downtimes may occur due to events such as hardware failures, software crashes, hacker attacks such as denial of service attacks, power outages where no back power is present, environment issues such as over temperature of servers due to non-functioning temperature control system of the data center etc.

Availability is usually presented as a percentage of uptime in a given year. For this, only the unplanned downtime is considered as a norm. Availability of the system is calculated as; minutes for a year minus total unplanned downtime in minutes divided by minutes for an year. Usually highly available systems have availability such as 99.9%, 99.99% and 99.999%. The first one has 8.76 hours unplanned downtime per year and the last has only 5.26 minutes per year.