By IT Brew Staff
less than 3 min read
Definition:
Highly available IT systems can meet a predetermined operational service level with a very small amount (if any) of downtime. For example, critical infrastructure in industries such as healthcare and telecommunications aims for “five nines” availability (i.e., available 99.999% of the time), which translates into just a few minutes of downtime per year. Autonomous vehicles, mission-critical Internet of Things (IoT) infrastructure, and electronic health records are all examples of technology that need the support of a high-availability network.
Other industries may not require that same level of availability, with users happy (at least in theory) with “four nines” (i.e., 99.99%) or “three nines” (99.9%) availability.
Building out a system with high availability may require:
- Hardware, data, and software redundancy
- Data replication in the event that a network cluster or device fails
- A failover system with a backup component in place
- A high fault tolerance that allows systems to keep operating despite failures
For those IT pros who want to facilitate high availability within their organizations, it’s critical to remove single points of failure, install tools that automatically detect infrastructure failures (or imminent infrastructure failures), and build out reliable failover and backup capabilities. Load balancing, or boosting service availability by spreading traffic across multiple servers in a network, can likewise prove essential, as can high-availability clusters, which can transfer workloads between devices in case of a failure.
