AWS Resilience Hub updates add AI-powered dependency insights and EKS support

These updates provide streaming infrastructure teams with automated tools to manage the complexity of distributed Kubernetes workloads. By using generative AI to map dependencies, platform engineers can identify single points of failure that often lead to service outages during regional traffic spikes. The ability to enforce uniform resilience policies via AWS Organizations ensures that fragmented engineering teams maintain consistent uptime standards. As streaming platforms scale globally, watch for how these automated insights reduce the mean time to recovery (MTTR) during large-scale live events.

AWS Resilience Hub sits within a growing category of cloud resilience and chaos engineering tools that streaming platforms increasingly rely on to maintain uptime during traffic surges. In March 2025, AWS announced that Resilience Hub had expanded its assessment capabilities to cover over 200 AWS resource types, including Amazon RDS, Amazon ElastiCache, and AWS Lambda, giving platform teams broader visibility into failure modes across the services that typically underpin streaming backends. The EKS label support announced in September 2026 extends that coverage into Kubernetes-native workloads, which now handle a significant share of video processing and delivery pipelines at companies like Netflix, Disney, and Warner Bros. Discovery.

On the business and competitive side, AWS has been positioning Resilience Hub as part of its broader Well-Architected Framework push to lock in enterprise platform teams. In June 2025, AWS introduced a new Resilience Hub pricing model based on the number of applications assessed rather than per-resource charges, a move aimed at reducing cost friction for large organizations running hundreds of microservices. Meanwhile, Microsoft Azure announced in May 2025 that its Azure Chaos Studio service had added integration with Azure Kubernetes Service for automated fault injection, directly competing with the EKS-focused resilience testing that AWS Resilience Hub now supports. Google Cloud has similarly expanded its resilience tooling, with Google Cloud announcing in April 2025 that its Application Resilience Assessment feature had reached general availability for GKE workloads.

From a technical standpoint, the generative AI dependency insights in AWS Resilience Hub updates represent a shift toward automated failure mode analysis that previously required manual architecture reviews. Gremlin published a 2025 State of Reliability report finding that 68% of platform engineering teams still relied on manual runbooks for failure scenario identification, suggesting significant room for practical AI applications to reduce mean time to detection. In the streaming-specific context, AWS re:Invent 2025 featured a session where Amazon Prime Video engineers described using Resilience Hub to validate failover paths for their global content delivery pipeline, demonstrating that the tool is already being applied to video workloads at scale. The cross-account policy sharing via AWS Organizations addresses a common pain point for streaming companies operating multiple AWS accounts for different regions or business units, where inconsistent resilience configurations have historically led to cascading failures during live events.