How we keep it up

When we talk about reliability, we mean carrier class reliability — a network that runs all day, every day, with no single point of failure and no excuses.

• No single points of failure. The network is redundant and diverse at every stage.

• Proactive monitoring of every point and service on the network. If any point runs into trouble, we know — and we act, any time day or night.

How do you build a carrier-class network? It’s not easy. It has to be designed from the ground up so that any single piece of equipment can fail without taking down the whole network. Individual switches may fail. Individual routers may fail. Individual backhaul links, from the tower to the NOC, may fail — but the network stays up.

All our routers have live standby backups. All switches are duplicated with automatic failover. All network paths from the tower onward are redundant, and it goes without saying that all network equipment has backup power. We have multiple redundant paths to the internet from our NOC.

None of that means anything without people watching over it. We have those people — you might never meet them or know their names, but they’re on the job around the clock, watching over the network.