Event-driven rendering architecture leveraging AWS Spot Instances across two availability zones with automatic interruption handling via EventBridge and Lambda. Users route through Route 53, CloudFront CDN, and ALB protected by WAF to reach EC2 Spot render workers and ECS Fargate tasks distributed across AZ-1 and AZ-2. When Spot interruptions occur, EventBridge triggers a Lambda Checkpoint Handler that persists job state to DynamoDB and ElastiCache, then requeues work via SQS for seamless failover to standby capacity. S3 events trigger Step Functions pipelines and Lambda handlers to orchestrate asset processing, while RDS Primary-Standby replication and DynamoDB ensure data consistency across zones. CloudWatch monitoring and CloudTrail auditing provide observability; IAM and Secrets Manager enforce least-privilege access. This pattern reduces compute costs by 70-90% using Spot while maintaining rendering SLA through event-driven checkpointing and multi-AZ resilience. Fork and customize this diagram on Diagrams.so to adapt checkpoint intervals, queue thresholds, or add additional event sources. Consider adding SNS dead-letter notifications for failed render jobs requiring manual intervention.