A financial services company operates a business-critical application hosted on Amazon EC2 instances behind an Application Load Balancer (ALB). These instances are managed by an Auto Scaling group distributed across multiple Availability Zones. On the first of each month at exactly midnight, the system experiences a dramatic slowdown as a heavy batch-processing job is triggered to run monthly financial reconciliations. During this window, CPU usage spikes to 100%, severely impacting application performance. What should a solutions architect recommend to ensure the application remains responsive and avoids outages during these predictable spikes?