Lessons from the AWS Outage and What They Mean for Healthcare Part 2: Reliability Isn’t Scale, It’s Design, Governance, and Intent

The AWS event last month provided a valuable reminder to me about the cloud that I wanted to share. When it comes to reliability in healthcare, the size of a cloud doesn’t matter. Reliability is engineered by experts through design, governance, and intent, not scale.

When the October 20 AWS outage rippled through the internet, the world saw how interdependent the digital ecosystem has become. Systems that were supposed to be regionally resilient struggled as dependencies stacked up. What the headlines missed was that many healthcare technology vendors found themselves in a critical gray zone. Their applications were technically online, but functionally unusable. In a clinical setting, that difference can mean losing access to systems when it is needed most.

The Overlooked Divide: Disaster Recovery vs. Operational Recovery

Most organizations prepare for disaster recovery, the plan for what happens when everything goes dark. However, healthcare leaders recognize that the majority of crises are not complete outages. There are slowdowns, partial failures, and cascading latency events. This is where operational recovery matters most. Disaster recovery is a blueprint. It defines where data lives, how replication occurs, and how long it takes to rebuild an environment. Operational recovery is a behavior. It’s how teams detect, decide, and act when systems degrade but do not fail. In healthcare, downtime is rarely absolute. A delay of a few seconds in order entry or a stalled imaging transfer can interrupt care as much as an outage. Operational recovery keeps those workflows moving when technology stumbles.

A Real Example: Restoring Services in Minutes, Not Hours

During the AWS incident, several independent software vendors serving hospitals across the U.S. experienced severe latency and service interruptions. One SaaS vendor, which was hosting its entire platform in AWS, joined a joint bridge call with CloudWave and the affected healthcare customer.

As both sides analyzed the situation, the CloudWave operations team recognized that waiting for AWS regional recovery or cross-region replication would take hours. Instead, the team proposed and executed a faster path: building and deploying a mirror configuration of the vendor’s environment inside CloudWave’s managed private cloud.

Because CloudWave was already hosting the customer’s Electronic Medical Record (EMR) system within its private cloud, along with the customer’s secure connectivity, identity integrations, and healthcare-specific networking requirements, our team was able to rapidly stand up the mirrored environment. Within a short window, the customer’s critical services were restored and accessible, even as the vendor’s SaaS offering in AWS remained unavailable.

This was not an improvised workaround. It was operational recovery in action—a disciplined process of restoring care-critical functionality, independent of hyperscaler status.

Why This Matters

In a hyperscale world, recovery is often viewed as an infrastructure challenge. In healthcare, it’s an imperative for patient safety. The best-engineered platform still requires people, intent, and governance to translate uptime metrics into clinical continuity.

A managed private cloud excels because it prioritizes clinical workflows when disruptions occur, not generic SLAs. It maintains full control and visibility across every layer of dependencies. It enables coordinated recovery actions among operations, security, and vendor teams in real-time.

The Lesson

For me and members of my CloudWave team, the AWS outage illustrated in real-time that reliability is never about footprint or fame. It’s about preparedness, purpose, and people. Healthcare organizations need partners who can act, not just wait, when global infrastructure falters.

When the next disruption happens, success in healthcare will not be measured by how fast a region recovers. Instead, it will be measured by how seamlessly safe and uninterrupted patient care is continued.

Mike Donahue
Chief Operating Officer
CloudWave.