Redundant Power, Network & Hard Drives
How do we achieve our multi-year average uptime for our VPS nodes? By taking critical aspects of our servers and making them redundant. Power and network are the usual culprits to downtime besides hardware failure or user-caused downtime. This means we have separate power feeds from different power grids feeding redundant power supplies in our servers. Should a PSU or power grid go down, servers would still stay online. This also goes for network; redundant 1Gbps network feeds from separate routing points in our network feed our servers. As for hardware, we have made our hard drives redundant in Raid-10, meaning up to two hard drives can fail at the same time and keep the server running.