A high-availability cluster is specifically designed to ensure that applications remain accessible with minimal downtime. This is achieved by having redundant nodes and automatic failover processes in place, so if one node fails, another node can take over with little to no disruption in service. The failover cluster configuration is the correct answer because it is built with redundancy and automatic failover in mind, which is essential for applications demanding continuous uptime. Load-balanced and horizontal scaling clusters are incorrect, as they are primarily used to distribute workloads and increase capacity rather than to provide failover capabilities. A single-node cluster is not a cluster by definition and provides no redundancy or failover support; hence, it is incorrect.