Continuity Planning
For hospitals, data centers, financial systems, and emergency services where any downtime is unacceptable. Find high-availability internet and power solutions at your address.
Continuous availability is an architecture standard that goes beyond high availability. Where high availability aims for 99.9% uptime (about 8.7 hours of downtime per year), continuous availability targets 99.999% — about 5 minutes per year — or higher. For internet and power, this means designing systems where no single failure can interrupt operations.
For data centers, continuous availability aligns with the Uptime Institute's Tier standards. Tier III (N+1 redundancy — one extra component beyond what is needed) achieves 99.982% uptime. Tier IV (2N+1 fully fault-tolerant redundancy) achieves 99.995% uptime. Both use industrial-grade diesel or natural gas generators sized for the full critical load, Battery Energy Storage Systems (BESS) for instantaneous bridge power, automatic transfer switches, and in the highest tier, dual utility feeds from separate substations so even grid infrastructure failure does not create a single point of failure.
For hospitals and healthcare facilities, NFPA 110 Level 1 requirements essentially mandate continuous availability for life-critical systems. Emergency power must restore within 10 seconds of grid failure. Facilities must carry 96 hours of on-site fuel. Generators must be tested weekly (inspection) and monthly (load test). Manufacturers such as Caterpillar, Cummins, Generac Industrial, and mtu by Rolls-Royce build the industrial-grade units used in these environments.
Microgrid technology is increasingly used to achieve continuous availability in complex facilities. A microgrid combines generators, BESS, solar, and utility feeds into a single managed system that can island (operate independently from the grid) during regional outages while automatically balancing load across all available sources.
Each layer eliminates a different failure mode. True continuous availability requires all layers.
High availability (HA) means 99.9% uptime — about 8.7 hours of downtime per year. Continuous availability (CA) means 99.999% or higher — about 5 minutes per year. CA requires true redundancy at every single point of failure. HA permits some single points of failure to exist, accepting brief interruptions as acceptable.
A complete continuous availability power stack includes: a large-scale UPS or BESS for instantaneous (sub-millisecond) failover, an automatic transfer switch (ATS) bringing generators online within 10 seconds, multiple standby generators in N+1 or 2N configuration, and ideally dual utility feeds from separate substations. Top manufacturers: Caterpillar, Cummins, mtu (Rolls-Royce), Generac Industrial.
A BESS is a large-scale battery bank (lithium-ion or similar technology) that stores electrical energy and can discharge instantly during a power event. Unlike a traditional UPS sized for minutes, a BESS can be sized for hours. It eliminates the 10-second gap between grid failure and generator startup — making power truly uninterrupted. BESS systems are now standard in Tier III and Tier IV data centers and increasingly used in hospitals.
A microgrid combines generators, BESS, solar, and utility feeds into a single managed system. During a grid outage, the microgrid islands — disconnecting from the utility and operating independently, balancing load across all available internal sources. Microgrids can sustain operations indefinitely as long as fuel or renewable input continues. They are increasingly deployed in hospitals, campuses, and data centers.
At minimum: two internet connections from two different providers on two different physical paths (fiber + LTE), with a router or SD-WAN platform that performs instant automatic failover with zero packet loss. Enterprise SD-WAN can manage three or more paths simultaneously, optimizing traffic across all available connections in real time.
Hospitals and NFPA Level 1 facilities, data centers (Tier III/IV), emergency dispatch centers, financial trading platforms, major e-commerce operations, and any operation where seconds of downtime directly causes patient harm or massive financial loss.