An ecommerce application uses PostgreSQL on an EC2 instance. Monthly sales events cause unpredictable spikes in database usage and connection issues. The company needs to maintain performance during these unpredictable increases in the most cost-effective way. Which solution resolves this issue?

Migrating to Amazon Aurora Serverless v2 is the most cost-effective solution for unpredictable spikes. Aurora Serverless v2 scales compute and memory capacity in fine-grained increments, instantly adapting to demand without over-provisioning, meaning you only pay for the resources consumed. Auto scaling for PostgreSQL on an EC2 instance is not a native feature for database scaling and would require complex, custom solutions for read replicas or sharding, which are difficult to implement and manage effectively for sudden spikes. Migrating to a larger Amazon RDS instance type addresses performance but is not cost-effective for unpredictable spikes as you would pay for the larger instance 24/7, even during low usage. Amazon Redshift is a data warehouse service optimized for analytical workloads, not transactional OLTP databases like the one described for an e-commerce application, making it an unsuitable choice.