Loading...
Quick Definition
High availability (HA) is a design approach that keeps systems operational with minimal downtime, typically by removing single points of failure through redundancy across zones or regions. HA improves reliability but adds infrastructure that must be balanced against cost.
High availability is the practice of designing systems to survive component failures without meaningful downtime. The method is redundancy: run multiple copies of everything across independent failure domains, typically availability zones, so the loss of any one machine, rack, or data center leaves the service running.
Availability is bought with money, and the price rises steeply at the top. Each additional nine of uptime roughly multiplies cost: a second zone doubles parts of the footprint, a second region can double it again. The financial discipline is matching the availability target to what the business actually loses during an outage, expressed as a service level objective rather than a reflexive demand for maximum uptime everywhere.
Example. A payments API runs three replicas across three zones behind a load balancer, because an hour of downtime costs real revenue. The internal reporting tool runs one replica, because nobody is harmed if it restarts at 2 a.m.
Review availability tiers per workload yearly; requirements drift and yesterday's critical system may be today's background tool. The Cloud Cost Management Guide covers balancing resilience and spend.
Roughly 8.8 hours of allowed downtime per year. Each extra nine cuts that tenfold and raises cost substantially.
Usually no. Multi-zone within one region survives most real failures. Multi-region is for the rare workloads that justify its cost.
Redundant capacity, cross-zone traffic, replicated storage, and standby databases all bill continuously, even when never used.