High Availability, Fault Tolerance and Disaster Recovery

High availability (HA)

High availability means ensuring agreed level of uptime, for higher than normal period.
Some disruptions are allowed, like re-logging in, some level of (recoverable) data loss etc.

Enables the system to continue operating properly in the event of failure of some (or more) components.
Often includes extra or duplicate components so that when one fails, other can keep running.

HA means maximizing uptime, and FT means operate through failure.
HA means you should be able to replace faulty server with working one, and FT means your frontend should continue working while backend is being replaced.
HA is less severe problem, FT is much complex.
HA is like changing a punctured tire of car, and FT is like extra engine in a plane.
Mission critical systems must be Fault Tolerant. While regular systems can only go till the High availability Zones

Set of policies, tools and procedures to enable recovery or continuation of vital components and infrastructures following a disaster.
Disaster recovery includes backups, moving things, people and knowledge to different locations and making recovery as smooth and fast as possible.