IOPS and Throughput Calculator

Formula: Throughput in MiB/s = IOPS × block size in KiB ÷ 1,024. IOPS needed = target MiB/s × 1,024 ÷ block KiB. Queue depth = IOPS × latency (Little's law). Volume maximums follow each provider's per-GiB rules at the size entered.

Worked example:3,000 IOPS with a 16 KiB average block size produces 46.875 MiB/s, which a 500 GiB gp3 volume meets at its 3,000 IOPS / 125 MiB/s baseline.

Common mistake or edge case:This is an arithmetic ceiling, not a benchmark. Queue depth, latency, access pattern, caching, and instance-level EBS or VM caps can lower observed throughput.

Additional official references:

• AWS EBS General Purpose SSD (gp3, gp2)
• AWS EBS Provisioned IOPS SSD (io2 Block Express, io1)
• Google Cloud Persistent Disk performance

Read the supporting guide and worked examples.

Frequently asked questions

How do I convert IOPS to MB/s?

Multiply IOPS by the block size in KiB and divide by 1,024 for MiB/s; multiply MiB/s by 1.048576 for decimal MB/s. 3,000 IOPS at 16 KiB is 46.875 MiB/s.

What are the gp3 limits?

gp3 includes 3,000 IOPS and 125 MiB/s, and can be provisioned up to 80,000 IOPS (500 IOPS per GiB) and 2,000 MiB/s (0.25 MiB/s per provisioned IOPS) on volumes from 1 GiB to 64 TiB, per AWS documentation.

Why do large I/Os use more EBS IOPS?

EBS SSD volumes count I/O in units of up to 256 KiB, so a 1 MiB operation counts as four IOPS. Sequential small I/Os may be merged into one.