An analytics company runs a data aggregation workload using Amazon EC2 instances. The application performs frequent and repetitive read operations on multiple Amazon RDS databases as part of a nightly batch process. The database read activity is contributing to performance bottlenecks. A solutions architect needs to reduce the load on the RDS databases and improve efficiency while maintaining high availability. What is the best solution?