Availability & Recovery — Network+

High availability (HA) is an architecture strategy that eliminates single points of failure so critical network services stay accessible despite hardware faults, software crashes, or maintenance windows. Common implementations include redundant power supplies, dual uplinks, NIC teaming, and clustering, where a standby node takes over automatically if the active node fails. The exam distinguishes HA from disaster recovery: HA prevents downtime through real-time redundancy, while disaster recovery restores operations after a major outage using backups and alternate sites. Uptime is expressed as a percentage — “five nines” (99.999%) allows roughly five minutes of downtime per year.

Redundancy is the practice of duplicating network components — links, power supplies, switches, or entire paths — so that no single failure brings down connectivity or services. On N10-009 it spans physical hardware (dual power supplies, bonded uplinks) and logical protocols. First Hop Redundancy Protocols such as HSRP, VRRP, and GLBP keep the default gateway available when a router fails, electing a standby device to assume a virtual IP and MAC address. The key exam distinction is that redundancy provides the spare capacity, while high availability is the measured outcome (uptime percentage). Redundant links still yield poor uptime if failover is slow, so the two work together rather than being interchangeable.

Load balancing distributes incoming client requests across a pool of servers so no single server bears the full workload, improving both throughput and fault tolerance. Common algorithms include round-robin (requests rotate through servers in order), least-connections (traffic goes to the server with the fewest active sessions), and weighted distribution (higher-capacity servers receive proportionally more traffic). The exam distinction from simple redundancy matters: redundancy keeps a standby resource idle until needed, while load balancing actively uses all servers at once. A load balancer also runs health checks, automatically removing an unresponsive server from rotation.

Disaster recovery (DR) is the documented set of plans, procedures, and infrastructure for restoring network services and data after a catastrophic failure, natural disaster, or major cyberattack. DR planning identifies backup sites — hot (fully operational and ready), warm (partially equipped), and cold (space only) — and assigns clear responsibilities so restoration proceeds without guesswork. The key exam distinction is RTO (Recovery Time Objective), the maximum acceptable downtime, versus RPO (Recovery Point Objective), the maximum acceptable data loss measured backward from the failure. DR governs both, while high availability focuses on preventing the outage in the first place — a common N10-009 trap.