Amazon Web Services (AWS) has announced major changes to its support network following the widespread outage that occurred on December 7. The company admitted that its previous approach to outage response and communication did not meet customer needs, leaving many in the dark about the status of their services.

After reviewing the incident, AWS outlined plans to redesign its support system and introduce an improved version of its service performance dashboard. The outage was triggered by an attempt to increase capacity on an internal service, leading to unexpected strains on the company’s internal network that handles critical functions like monitoring, DNS, and authorization. This in turn overwhelmed systems responsible for managing communication between AWS’s core and internal networks.

A wide range of customers—from financial platforms to consumer electronics and streaming services—were impacted, as well as Amazon’s own e-commerce site. The disruption was centered in the northern Virginia region, home to AWS’s largest data center cluster. While the outage lasted from 7:30 a.m. to 2:22 p.m. PST, it took until nearly 7:00 p.m. to fully restore normal response times for all services.

To prevent similar incidents in the future, AWS will implement an active-active multi-region support architecture. This approach distributes workloads and support across several regions simultaneously, ensuring service continuity even if one region suffers an outage. While more complex to build and maintain, this system drastically reduces single points of failure.

Mattias Andersson, a senior architect specializing in cloud training, explained that the primary challenge with active-active systems is keeping data synchronized across multiple regions. However, he praised AWS’s decision, citing improved resilience and reliability for customers.

The outage also revealed that AWS’s communication channels, especially its Service Health Dashboard, were inadequate during the incident. AWS plans to upgrade the dashboard in the coming months for better transparency during outages.

Industry experts advised customers to educate themselves about cloud infrastructure and evaluate how their workloads are architected for cloud resilience. Achieving 100% uptime is expensive and generally unnecessary, but understanding the cloud’s strengths and limitations helps organizations mitigate risks.

AWS recommends that IT teams make use of best practices frameworks to ensure their systems are designed for reliability and availability in the cloud.