High-Availability vs Fault-Tolerance vs Disaster Recover

High-Availability

• when it fail, its component can replace as soon as possible
• = maximize the online time of a service/uptime = minimize downtime
• measure by available time (e.g. 99.999% = 5.26 minute/ year downtime)
• allow interuption

Fault-Tolerance

• more than high-availability
• it is a property that enables a system to continue operating properly in the event of the failure of some of its components
• actively connected to multiple servers
• minimize interruption occur even with failure
• can cost much more

Disaster Recover

• a set of policies, tools and procedures to enable the recovery or continuation of vital technology infrastructure & systems following a natural or human-induced disaster
• pre-planning + (automated) DR process
• back-up plan when disaster
• multiple site of backups