A company is migrating a distributed application to AWS. The legacy platform has a primary server that coordinates jobs across multiple compute nodes. The application serves variable workloads. The company wants to modernize the application to maximize resiliency and scalability. How should a solutions architect design the architecture to meet these requirements?

Using Amazon SQS as a destination for jobs decouples the job producers from the consumers (compute nodes), enhancing resiliency by buffering requests during peak loads and ensuring no jobs are lost if consumers are unavailable. Implementing compute nodes with EC2 instances in an Auto Scaling group provides scalability. Configuring Auto Scaling based on the SQS queue size directly links the scaling of compute resources to the actual workload demand, ensuring efficient resource utilization and responsiveness to variable workloads. Incorrect options: Scheduled scaling for EC2 Auto Scaling is less dynamic than queue-based scaling and wouldn't react optimally to variable, unpredictable workloads. Using CloudTrail as a destination for jobs is incorrect; CloudTrail is for logging API activity, not for queuing application jobs. EventBridge (CloudWatch Events) is a serverless event bus for routing events, not a message queue for application jobs. Scaling based on primary server load or compute node load is less direct than scaling based on the pending work in the queue.