A single-writer Amazon Aurora PostgreSQL cluster experienced over 90% utilization for 11 minutes when a weekly automated report ran, causing user-facing slowdowns. How should the administrator prevent weekly performance degradation caused by the report?

Adding a reader instance to the Aurora PostgreSQL cluster and directing the report to it offloads the read-heavy workload from the primary writer instance. This prevents the report from contending for resources with user-facing applications, which continue to use the writer instance for both reads and writes, thus resolving the performance degradation. Scaling the existing DB instance might temporarily alleviate the issue but is a more expensive solution and doesn't address the root cause of a read-heavy workload impacting the writer. Adding another writer instance is incorrect because Aurora PostgreSQL clusters have only one writer instance; additional instances are always readers. Enabling auto scaling for the DB cluster is not applicable to Aurora instances in the same way it is for EC2 instances; Aurora handles scaling storage automatically, but compute scaling for a single instance requires manual modification or a different architectural approach like reader instances.