NG911 Resiliency and Redundancy: Eight Steps to a Resilient 911 Network

Halfway through building the Brooklyn Bridge, Washington Roebling learned that a shady contractor had slipped 221 tons of rejected, brittle wire into its main cables. Due to how they were spun, pulling the bad strands out was practically impossible. The bridge opened in 1883 and, to this day, it carries traffic across the East River, with the rejected wires still sealed inside. His father, John Roebling, had designed the cables to carry six times their expected load, so Washington ordered 150 extra wires for each cable and kept building. 

Its secret is the safety margin John Roebling built into the design, wide enough that when defective wire from a contractor turned up years later, the bridge could absorb the flaw. Every emergency communications center (ECC) could build its 911 network the same way. The centers that do are the ones still standing when lightning, ransomware, or a hurricane exploits the weak points in the network.

Most ECC leaders believe they already have that margin of safety. The contract lists diverse paths, there’s a backup line, and even a disaster recovery plan. Yet, redundancy proves itself only under load, and by the time the load arrives, the window to fix it has closed.

Below, we define what resiliency and redundancy mean in NG911, show where the gaps hide, walk through three resilient agencies, and close with an eight-point checklist you could put into practice today.

Where the Gaps Hide

In the spring of 2023, a tornado in the upper Midwest cut a single fiber line. Nobody knew that strand was the lone point of failure for the carrier serving four public safety answering points (PSAPs). Calls failed over to a secondary path, and the backup worked. However, its provider had no network operations center and no 24/7 support, so four centers ran on an unwatched line for two days until crews repaired the fiber.

Look closely, and the weak “wire” in that story was designed during the contract phase. Those four centers were exposed years before the tornado touched down.

Resiliency is the practiced ability to keep answering when the primary fails and the spare takes the load. In order to protect your network, you account for redundancy. In other words, a second route into the Emergency Services IP Network (ESInet), a second site, and a second system that shares no components with the first. 

Done well, NG911 delivers four layers of redundancy that legacy selective routers never could:

• ESInet path diversity, with more than one route enter and deliver on the network.
• Policy-based routing that reroutes calls automatically on overflow or outage.
• Geodiverse location and routing functions (ECRF and LVF) that share no single point of failure.
• Dynamic transfer between centers when one is overwhelmed.

The Midwest outage taught us three valuable lessons. First, diverse paths sometimes converge on one strand of fiber a state away, invisible until it becomes an issue. Second, an unmonitored backup leaves dispatchers alone on the line. And third, resilience is implemented long before any storm.

The same pattern plays out at national scale. A few years back, one carrier’s network failure cascaded across as many as 39 states and disrupted 911 for nearly 37 hours. About 17 million customers lacked reliable access to 911, and at least 886 calls were never delivered.

Interoperability adds its own seams. Neighboring ECCs served by different ESInet providers can’t always transfer calls natively. Implementations of the NENA i3 standard vary by version, which complicates interconnection. Mismatched GIS boundary data between jurisdictions misroute calls. And when no mutual-aid routing is provisioned in advance, directors are forced to improvise in the middle of the event.

Three Tests in the Field

Lightning in Clark County, Indiana. When state law required Clark County 911 to consolidate six PSAPs into one primary and one secondary, Director Brad Meixell went looking for a backup from a provider independent of his primary system. The county chose MEVO, INDIGITAL’s disaster recovery platform. On April 8, 2015, lightning damaged the main 911 system. Meixell switched call delivery, and calls routed automatically to MEVO.

“We have never had any problems with it,” Meixell said. “It has always worked.”

Ransomware in Rock Island County, Illinois. At about 8 a.m. CDT on March 24, 2021, ransomware disrupted the call handling system serving the county. Routing policies written long before the attack sent unanswered 911 calls to MEVO as ring-no-answer timers expired, and the first call arrived at 8:04 a.m. Over the next six days, four PSAPs handled 1,482 911 calls and nearly 6,000 administrative calls on MEVO while INDIGITAL, county IT, the State of Illinois, and security consultants rebuilt the primary system. Primary call handling returned March 30.

Back-to-back hurricanes, 2024. After Hurricane Helene knocked out power and communications across Georgia in late September, INDIGITAL deployed MEVO Anywhere Kits and Starlink satellite units. Working with the Alabama 911 Board, Alabama PSAPs, and the Georgia Emergency Management Agency, the team reconnected nine 911 centers, including those in Burke, Tift, and Johnson counties. Two weeks later, a Florida response team staged statewide ahead of Hurricane Milton with more than 20 Starlink units, while INDIGITAL’s network operations center held twice-daily calls with every Florida PSAP.

That work earned INDIGITAL Frost & Sullivan’s 2026 U.S. Product Leadership Recognition in Disaster Recovery Solutions as the category’s sole recipient.

Not every center can change networks before its next emergency. ENCORE, INDIGITAL’s disaster recovery service for PSAPs on any network, reroutes 911 delivery with ALI display, mapping, and call recording. It reached Holmes and Jackson counties in Florida after a July 2024 outage.

While technology is vital, it’s the people who own the response. Kristi Stamnes, director of Baldwin County 911, put it plainly: “You don’t realize exactly what kind of partnership you have until something goes wrong, and then you realize that they’re one of the legs that supports us.”

Your Eight-Point Resilience Checklist

1. Map every single point of failure: carrier, ingress, power, backhaul, and GIS.
2. Confirm true path diversity, with separate providers behind separate circuits.
3. Run a live failover drill this season, beyond a tabletop exercise.
4. Verify that your backup path has a network operations center and 24/7 support.
5. Sign cross-jurisdiction transfer and mutual-aid routing agreements before you need them.
6. Check GIS boundary accuracy and keep ECRF and LVF data current.
7. Establish a communications fallback for when CAD or admin lines fail and the call path holds.
8. Schedule a recurring resilience review, owned by a named person.

The Bottom Line

In 1884, P.T. Barnum marched 21 elephants across the Brooklyn Bridge to show the city it would hold. The extra wires were working. It’s no different from how the agencies in Clark County, Rock Island County, and across the Southeast kept their network online. 

Your next quiet season is the time to find your own weak wire. Talk with INDIGITAL about a resilience review for your center, or explore MEVO disaster recovery at indigital.net/services/mevo.

911 Is Our Calling.™