If you still run gp2 EBS volumes, migrating to gp3 is one of the highest-return, lowest-risk changes you can make on AWS. It is cheaper, it is faster, and the switch happens with no downtime. Here is the math and the method.
The price difference
gp3 storage is priced roughly 20% lower per GB-month than gp2. Crucially, gp3 decouples performance from capacity: every gp3 volume includes a baseline of 3,000 IOPS and 125 MB/s for free, regardless of size. With gp2, IOPS scale with volume size, so teams often over-provision capacity just to buy performance.
Why gp3 is usually faster too
A small gp2 volume (say 100 GB) only gets 300 baseline IOPS. The equivalent gp3 volume starts at 3,000 IOPS. For most databases and general workloads, gp3 is a straight upgrade on both cost and performance.
The one thing to check first
gp2 volumes larger than ~1 TB can exceed 3,000 baseline IOPS on gp2. Before migrating, confirm the volume's actual IOPS and throughput usage. If a volume genuinely needs more than the gp3 baseline, you can provision extra gp3 IOPS/throughput — and it is often still cheaper than the gp2 equivalent. Check real usage, don't assume.
How to migrate (no downtime)
- Select the volume in the EC2 console → Modify Volume.
- Change the type from gp2 to gp3. Keep size the same.
- If needed, set IOPS/throughput above the 3,000 / 125 baseline to match real usage.
- Apply. The change happens live; the volume stays attached and in use.
Do it across every account at once
The hard part is not migrating one volume — it is finding all the gp2 volumes across every account and region, checking their real IOPS, and tracking the savings. CloudMonitor surfaces every gp2 volume as a ranked recommendation with the projected dollar saving and the exact steps. It is one line in the broader AWS cost-reduction playbook, and the cost optimizer handles it automatically.