Difference between revisions of "High availability"
| (One intermediate revision by the same user not shown) | |||
| Line 1: | Line 1: | ||
| − | [[High availability]] (also known by its acronym, [[HA]]; hereinafter, ''HA'') is the feature of a system that ensures higher operational performance such as uptime than a similar system without that feature would allow for. | + | [[High availability]] (also known by its acronym, [[HA]]; hereinafter, ''HA'') is the feature of a system that ensures higher operational performance such as [[uptime]] than a similar system without that feature would allow for. |
| − | + | In other words, ''HA'' is the property of a system to have a higher [[uptime]] than an identical system that does not use high availability tools and techniques. No system and no part of a system can be completely protected from the threat of abnormal operation or an emergency. ''HA'' can be described as the continued provision of services by the system at some "healthy" level when a certain part of it fails, while simultaneously recovering the very part that suffered from the failure. High-availability tools include redundant parts that are ready to take over the role of primary parts, monitoring devices to detect failures, and control devices that [[fencing]] non-working parts and redirect requests to working ones. The requirement for a “good”, albeit emergency, state distinguishes ''HA'' from the concept of [[failure tolerance]], which seeks to ensure that the average user of the system does not notice the failure of part of it. | |
Latest revision as of 03:10, 5 August 2023
High availability (also known by its acronym, HA; hereinafter, HA) is the feature of a system that ensures higher operational performance such as uptime than a similar system without that feature would allow for.
In other words, HA is the property of a system to have a higher uptime than an identical system that does not use high availability tools and techniques. No system and no part of a system can be completely protected from the threat of abnormal operation or an emergency. HA can be described as the continued provision of services by the system at some "healthy" level when a certain part of it fails, while simultaneously recovering the very part that suffered from the failure. High-availability tools include redundant parts that are ready to take over the role of primary parts, monitoring devices to detect failures, and control devices that fencing non-working parts and redirect requests to working ones. The requirement for a “good”, albeit emergency, state distinguishes HA from the concept of failure tolerance, which seeks to ensure that the average user of the system does not notice the failure of part of it.
