3.4 - High availability and redundancy
Network+ N10-009 objective 3.4 covers designing networks that stay up. An active-active high-availability configuration runs both nodes actively sharing the load at once, unlike active-passive where one stands by. Load balancing spreads incoming requests across multiple servers for scale and resilience. NIC teaming bonds two network adapters on a host so traffic survives one adapter failing. Clustering coordinates multiple servers so they act as one highly available system, failing over automatically. You should also know first-hop redundancy protocols like VRRP and HSRP, redundant power and links, and diverse paths. Expect scenario questions that describe two nodes sharing load, spreading requests, surviving a NIC failure or grouping servers into one resilient unit, and ask which high-availability technique - active-active, load balancing, NIC teaming or clustering - is being described.
Both nodes share load at once = active-active. Spread requests across servers = load balancing. Bond two NICs to survive one failing = NIC teaming. Servers acting as one HA system = clustering.
Practice questions
1. Which HA configuration runs both nodes actively sharing the load at once?
- Active-passive
- Active-active (correct answer)
- Cold standby
- N+1 spare
Active-active runs both nodes live, sharing load and giving failover. Active-passive keeps a standby idle until the primary fails; cold standby is powered off.
2. Which technique spreads incoming requests across multiple servers for scale?
- NIC teaming
- Load balancing (correct answer)
- Clustering
- Snapshotting
Load balancing distributes client requests across a server pool (round-robin, least-connections). NIC teaming bonds a host's NICs; clustering coordinates nodes for HA.
3. Which host feature bonds two NICs so traffic survives one adapter failing?
- Port mirroring
- Clustering
- NIC teaming (correct answer)
- Load balancing
NIC teaming groups a host's network adapters for redundancy and/or aggregated bandwidth, so one NIC or cable failing does not drop connectivity.
4. Which approach coordinates multiple servers so they act as one highly available system?
- Port mirroring
- Clustering (correct answer)
- Snapshotting
- Storm control
Clustering links nodes to share state and fail over so the service stays available if a node dies. Snapshots capture state; port mirroring copies traffic.
5. Which sampling technology exports a subset of packets for scalable flow visibility?
- Syslog
- SNMP
- sFlow (correct answer)
- NTP
sFlow samples packets at a set rate to scale on high-speed links, trading full accuracy for lower load. NetFlow records full flows; SNMP polls counters.
6. Which HA design keeps a standby node idle until the active node fails?
- Active-passive (correct answer)
- Active-active
- Round-robin
- Anycast
Active-passive holds a standby ready but idle, cutting over on failure. Active-active runs both nodes sharing load; round-robin is a balancing method.