High availability- explanation

High availability is a term used to describe a system or a service which is ready to be used, therefore “available”, almost all the time. Availability refers to the situation when a user has access to and can use the resources of a system, service or a device, either to work, add new elements or use the databases. If user cannot do any of those, it is said they are unavailable.

The time when system is not available is also called downtime. High availability ensures that the required level of operational performance is met. There is a scale of availability listing the percentage of overall times when the system is unavailable. In the case of high availability this “almost” is a matter of about 5 minutes a year when the system is unavailable. This is 99,999 percentage (five nines) of availability. There are two types of the periods when a system is unavailable. One is scheduled and the other unscheduled downtime. The first occurs when e.g. a new software is installed and a configuration changes or system reboots are necessary. During these, the system is also unavailable, but they are planned actions, that is why they are called scheduled downtime. The unscheduled downtime is caused by accidents, such as software and hardware failures or blackouts. The damages and losses caused by those unscheduled downtimes can multiply when unavailability increases. Whereas, the scheduled downtime for some systems will not impact the functioning of the whole systems, for example when in an office they occur at night, because the people working there are at home and do not use the system. Also, there is a relation between complexity of a system and its availability. The systems that consist of fewer components tend to have higher availability than the more complex ones. This is because there are more elements that are threatened by potential failures.