Your Guide to AWS Cloud Resilience Sessions at re:Invent 2026

Resilience is foundational to keeping what you build on AWS running. It matters to your customers, your stakeholders, and to your business; and re:Invent 2026 has a Cloud Resilience session to help you strengthen your approach. This year’s lineup spans breakout sessions, chalk talks, code talks, workshops, and lightning talks and we’ve organized them below to help you plan your week. Beyond the sessions, come find us at the expo hall at the Resilience and Cloud Operations booth in the AWS Village for hands-on demos, live conversations with resilience experts, and a first look at some of the latest launches and innovations in resilience. And for a more personal setting, join one of two Resilience Meetups for an open, interactive conversation with AWS resilience experts and fellow practitioners. Bring your toughest questions, share what’s working, and learn from others who are tackling similar challenges.

• Resilience Meetup 1 — Use AI to Take Your Resilience to the Next Level (COP370) [Tuesday, Dec 1, at 1 pm]
• Resilience Meetup 2 — Resilience Strategies And Insights (MUP314) [Wednesday, Dec 2, at 2 pm]

Assess and Strengthen Your Resilience Posture

Understanding where you stand is the first step to closing resilience gaps. These sessions show you how to map dependencies, assess your posture, and get actionable recommendations — before an incident does it for you.

Operationalizing resilience with AWS Resilience Hub (COP367) Learn how to continuously assess and validate the operational resilience of your end-to-end business services using AWS Resilience Hub. See how GenAI agents automatically discover and map your application dependencies to give you a holistic picture of your resilience posture. Get actionable recommendations to close gaps, generate regulatory compliance artifacts to meet your compliance needs, and put it all to the test with end-to-end resilience testing to prove it.

Build resilient apps on AWS: AI-driven resilience at any scale (COP316) Most teams discover resilience gaps during outages, not before. In this chalk talk, walk through the next generation of AWS Resilience Hub to map user journeys, define recovery objectives, identify dependencies, and run AI-powered failure mode assessments. Explore how to get prescriptive HA and DR recommendations and automate your DR mechanisms with meaningful tests. Leave with a step-by-step approach to strengthen the resilience posture of your critical applications, from identifying their business impact all the way to testing and automating their recovery.

Resilience starts where your assumptions end, feat. Autodesk (COP326) The challenge most companies face is knowing where they stand with resilience, being willing to validate their assumptions safely, and being disciplined enough to continuously improve without slowing delivery. In this breakout session, walk through Autodesk’s journey with resilience testing and automated recovery failover as proof that investing in validation pays off. Discover how the next-generation AWS Resilience Hub handles the tedious work — mapping dependencies, assessing posture, and preconfiguring resilience tests — leaving your teams to focus on closing the gap between the system you think you operate and the one you actually do.

Find failure modes faster: AI-powered resilience assessments (ARC307) Manual resilience reviews are slow, inconsistent, and routinely miss failure modes that don’t surface until an incident occurs. In this workshop, you’ll use the generative AI-driven assessment engine in AWS Resilience Hub to automatically discover dependencies across AWS services, internal endpoints, and third-party systems. You’ll evaluate workloads against the Well-Architected Framework and your own defined resilience policies including availability SLOs, RTO, and RPO targets, then receive prioritized, actionable recommendations tied to business impact. Working with a reference architecture, you’ll build a complete end-to-end assessment and leave with a repeatable framework you can run against any workload as your architecture evolves. Bring your laptop to participate.

Disaster Recovery and Failover

When disruption hits, recovery needs to be sequenced, tested, and repeatable. These sessions take you deep into orchestrating failover and failback across your full stack.

Recover and prove your full stack without disrupting production (COP334) Meeting your recovery time and point objectives is non-negotiable. In this lightning talk, see AWS Elastic Disaster Recovery replicate servers into low-cost staging and recover the full stack in the right order, in minutes. Learn how to run non-disruptive recovery drills against your own environment and prove your RTOs before an incident does it for you.

Orchestrate failover and failback for effective disaster recovery (COP314) Failing over one app is easy — orchestrating failover and failback across interdependent apps is where DR plans break. In this code talk, live-code full-stack Region failover using Amazon Application Recovery Controller (ARC) Region switch: building sequenced workflows so databases switch before app tiers, validating steps with health checks, rehearsing with practice runs, and safely reversing the sequence for failback. Leave ready to replace your paper DR plan with coded, repeatable recovery and measured RTO to prove it.

Orchestrate multi-layer disaster recovery for critical workloads (COP321) Your AZ recovery plan and your Region failover plan were probably built by different teams, tested at different times, and have never run together. Explore how to build a unified recovery framework with Amazon Application Recovery Controller — from zonal autoshift for rapid single-AZ mitigation to Region switch for cross-Region failover. Then discover how to run automated resilience tests that continuously validate your recovery posture before a real event forces the question.

Recovering AWS Graviton workloads with orchestrated disaster recovery (COP368) You moved to AWS Graviton for better price-performance, but is your disaster recovery keeping pace? In this chalk talk, whiteboard with disaster recovery experts how AWS Elastic Disaster Recovery (DRS) now protects Graviton-based (arm64) Linux servers and recovers them onto Graviton instances — keeping the architecture intact end to end with nothing extra to configure. Design a recovery plan that brings your application back tier by tier, in the right order, with wait times and critical-server gates. Bring your toughest arm64 and recovery-order questions, and leave with a DR design you can drill on Monday.

Test, Validate, and Prove Your Resilience

Your recovery plan is only as good as the architecture backing it — and the testing that proves it works. These sessions show you how to build for failure, inject faults, and walk away with evidence that your systems recover the way you designed them to.

From pod to multi-region: Resilient workloads on Amazon EKS (COP313) Building resilient workloads on Amazon EKS requires thinking in layers — from pod-level reliability to multi-region failover. In this chalk talk, explore each layer: Pod Disruption Budgets and Topology Spread Constraints, Karpenter and EKS Auto Mode for fast node recovery, EFS vs EBS for stateful workloads, AWS Backup for cross-region protection, and Route 53 with ARC for region switching. Debate active-active vs. active-passive tradeoffs and validate RTO/RPO targets with Resilience Hub.

Learnings from resilience testing in AWS (COP341) How does AWS test its own resilience? In this lightning talk, learn about the resilience testing practices used inside AWS — including the Game Day Region, why it exists, and how it’s used. Explore the types of tests AWS runs, the philosophy behind them, and takeaways you can apply to your own resilience testing program.

Break it to prove It: Fault-isolated architecture built to recover (ARC309) Your recovery plan is only as good as the architecture backing it. This workshop presents single-Region and multi-Region resilience patterns built along fault isolation boundaries, where failure scope dictates which recovery mechanism to engage. Using Amazon Application Recovery Controller (ARC), you architect for discrete blast radii, then validate each boundary through fault injection with AWS Fault Injection Service. Observability surfaces the failure, alarms classify scope, and automated recovery executes without intervention. The result: proof that your architecture recovers the way you designed it to. Bring your laptop to participate

Start planning your week

These sessions are your starting point for learning about resilience at re:Invent 2026. Search “resilience” in the re:Invent session catalog to discover the full list of over 200 sessions designed to help you build more resiliently on AWS. Download the AWS Events mobile app on iOS or Android to build your agenda. Reserved seating fills up fast, so claim your spots early.

See you in Las Vegas!