Calculate GCP Persistent Disk IOPS, throughput, and monthly costs by disk type.
Last verified: May 2026
| Type | Price | IOPS | Throughput | Use Case |
|---|---|---|---|---|
| Standard | $0.040/GB | 0.75 R / 1.5 W per GB | 0.12 MB/s per GB | Bulk storage, logs, backups |
| BalancedSelected | $0.100/GB | 6 R/W per GB | 0.28 MB/s per GB | General workloads, boot disks |
| SSD | $0.170/GB | 30 R/W per GB | 0.48 MB/s per GB | Databases, high IOPS apps |
| Extreme | $0.125/GB + $0.006/IOPS | Up to 120,000 (provisioned) | Up to 2,200 MB/s | SAP HANA, large databases |
Output will appear here...The GCP Persistent Disk Calculator determines the IOPS, throughput, and monthly cost for Google Cloud Persistent Disks based on disk type and size. It covers pd-standard (HDD), pd-balanced (SSD), pd-ssd, pd-extreme, and Hyperdisk options, each with different performance profiles that scale with capacity. Enter your desired disk size to see the resulting IOPS and throughput limits alongside the monthly cost, or specify your performance targets to find the minimum disk size required. This tool is essential for sizing boot disks, database volumes, and high-performance storage for Compute Engine VMs and GKE nodes.
pd-ssd provides 30 read IOPS and 30 write IOPS per GB, up to the per-volume maximum of 100,000 read and 100,000 write IOPS. So a 500 GB pd-ssd disk delivers 15,000 IOPS. The calculator shows the exact IOPS and throughput for any size you enter.
pd-extreme lets you provision up to 120,000 IOPS independently of disk size, unlike other disk types where IOPS scale with capacity. It is designed for the most demanding database workloads such as SAP HANA, Oracle, and SQL Server on large Compute Engine VMs. The calculator shows the separate IOPS provisioning cost.
Your PostgreSQL on Compute Engine needs 15,000 IOPS sustained. Using pd-ssd, that's a 500 GB disk at $85/month — but the workload only uses 80 GB of actual data. With Hyperdisk Balanced, you provision 100 GB capacity ($10/month) + 12,000 IOPS above the 3,000 free baseline ($60/month) + 240 MB/s throughput above the 140 MB/s baseline ($4/month) = $74/month. Same performance, $11/month savings, and you can scale IOPS independently when the database grows.
The calculator computes monthly disk cost as capacity × per-GB rate by disk type, plus (for Hyperdisk and pd-extreme) provisioned IOPS × per-IOPS rate and provisioned throughput × per-MBps rate. Inversely, given a target IOPS, it derives the minimum disk size for capacity-tied disk types (pd-ssd: 30 IOPS/GB, pd-balanced: 6 IOPS/GB) so you can right-size from a performance requirement.
Hyperdisk is GCP's answer to AWS gp3 / Azure Premium SSD v2 — it decouples capacity from performance, letting you provision IOPS and throughput independently. For most general-purpose workloads in 2026, Hyperdisk Balanced is cheaper than the equivalent pd-ssd configuration because you're not forced to over-provision capacity to get IOPS.
pd-ssd's 30 IOPS/GB scaling means a 100 GB pd-ssd disk only delivers 3,000 IOPS — barely enough for a small database. Either size up to a 333 GB disk for 10,000 IOPS, switch to Hyperdisk Balanced (which provides 3,000 IOPS at the baseline regardless of size), or expect IO-bound workloads to suffer.
Boot disks default to pd-balanced in newer machine types, but legacy templates still use pd-standard. pd-standard's max IOPS (7,500/disk) and throughput limits often bottleneck modern VMs. Always check what disk type the boot disk actually got — it's an easy 30%+ performance win to upgrade.
Was this tool helpful?
Disclaimer: This tool runs entirely in your browser. No data is sent to our servers. Always verify outputs before using them in production. AWS, Azure, and GCP are trademarks of their respective owners.