Question

A SysOps administrator manages a self-hosted PostgreSQL database on an EC2 instance. Users report slow query performance during peak hours. CloudWatch metrics show the attached General Purpose SSD (gp2) volume consistently hitting its burst balance depletion, and VolumeReadOps/VolumeWriteOps indicate a small-random-I/O workload requiring a sustained 12,000 IOPS at 250 GiB of stored data. The team wants to eliminate the throttling while keeping costs as low as possible. Which action best meets these requirements?

Accepted Answer

Migrate the volume to gp3 and provision 12,000 IOPS and the required throughput independently of the volume size

Answer

Increase the gp2 volume size to 4,000 GiB so the baseline IOPS reaches 12,000 (3 IOPS per GiB)

Answer

Migrate the volume to Provisioned IOPS SSD (io2 Block Express) and provision 12,000 IOPS

Answer

Migrate the volume to Throughput Optimized HDD (st1) to support the sustained workload