Fault Tolerance

Correct Answer: Load balancing with multiple servers

Load balancing distributes network or application traffic across various servers, which prevents any one server from becoming the bottleneck and allows optimum resource utilization, minimization of latency, and increased capacity. This approach provides fault tolerance- if one server goes down, the load balancer will automatically start routing traffic to the remaining online servers. The other answers are not as effective: daily data backups can protect against data loss but not server downtime; increasing RAM and CPU on the primary server doesn't protect against possible server failure; and a RAID-0 configuration provides no redundancy and therefore no protection against server failure.

Correct Answer: Implementing redundant switches

To prevent a single-point-of-failure on a network switch, the most effective strategy is to implement redundant switches. Redundancy, in this case, means having an extra backup switch (or switches) that can take over when the primary switch fails, thereby ensuring that the network does not go down. On the other hand, 'Upgrading switch firmware regularly', 'Implementing Quality of Service (QoS)', and 'Using patch management' are all good management tactics to maintain the health of the system and enhance security, but they do not directly prevent a single point of failure, which is primarily about ensuring physical or logical backup of key resources. Hence, their impact on preventing a single-point-of-failure is indirect and less substantial than implementing redundancy.

Correct Answer: Backup ISP connection with automated failover

The best configuration to ensure remote access to a network remains operational even if the main ISP (Internet Service Provider) connection fails is to have a 'Backup ISP connection with automated failover'. This setup serves as a contingency plan, automatically taking over internet services once the primary connection becomes unavailable, thereby ensuring a business' remote access capabilities remain undisturbed. The alternative options are not incorrect methods to handle network issues, but they do not directly address the specified problem. 'Multiprotocol Label Switching (MPLS)' is a type of data-carrying technique for high-performance networks but doesn't necessarily provide continuity during an ISP failure. 'Virtual Private Network (VPN) connections' provide secure access over a public network to an organization's resources but do not ensure business continuity if the ISP connection fails. Finally, 'Quality of Service (QoS) implementation' helps prioritize network traffic, but does not assure accessibility during main ISP connection failures.