Automated Student Data Pipeline with Kubernetes Scaling

Design a Kubernetes-based microservices architecture for a student information system that can dynamically scale based on enrollment periods. Create a Python Flask application that uses pandas for data processing, with Terraform configurations to deploy infrastructure. Implement horizontal pod autoscaling that responds to peak registration times, with automated monitoring using Prometheus and Grafana dashboards. Include robust error handling for data migrations and ensure zero-downtime deployments during critical academic calendar transitions.