Network-based block storage for Compute Engine VMs with automatic replication, snapshots, and multiple performance tiers.
What is Google Cloud Persistent Disk?
Persistent Disk is an established block storage service for Google Compute Engine VMs. It works as “network block devices”: the physical storage isn’t directly attached to the VM but is provided over the network and managed by Google. Data persists when VMs are stopped or deleted, and disks can be moved between VMs or attached to multiple VMs simultaneously in read-only mode.
A core feature is automatic replication within a zone, protecting data from hardware failures. Regional Persistent Disks synchronously replicate across two zones within a region for higher availability. Incremental snapshots enable efficient backups that can be created even from running VMs and restored in any region for disaster recovery.
Persistent Disk offers four performance tiers: Standard HDD (pd-standard) for cost-effective sequential workloads, Balanced SSD (pd-balanced) for general production workloads, Performance SSD (pd-ssd) for I/O-intensive applications, and Extreme SSD (pd-extreme) with provisionable IOPS for demanding database workloads on supported machine types. A single disk can be up to 64 TiB, and a VM can attach multiple disks totaling up to 257 TiB. Encryption-at-rest is enabled by default, with optional customer-managed encryption for additional control.
Important: For the latest machine series, Google now primarily recommends Hyperdisk, Google’s newer block storage generation with performance configurable independently of capacity. Persistent Disk remains available in parallel and continues to be supported on many machine types, but is not compatible with every new machine series.
Common Use Cases
VM Boot Disks for Compute Engine
Persistent Disk serves as standard boot storage for Compute Engine VMs. Boot disks contain the operating system and can be snapshot-cloned for rapid VM cloning. Balanced SSD offers a good price-performance ratio for boot disks, while Standard HDD suffices for development and test environments.
Database Storage for MySQL, PostgreSQL, MongoDB
Persistent Disk is suitable for relational and NoSQL databases. Balanced SSD provides sufficient performance for medium-sized databases, while Performance or Extreme SSD is used for workloads with higher IOPS requirements. Regional Persistent Disks provide additional availability for critical databases.
Persistent Storage for Container Workloads
Google Kubernetes Engine (GKE) uses Persistent Disk for PersistentVolumes. Stateful sets like databases or message queues can dynamically provision Persistent Disks via Kubernetes volume claims.
Snapshot-based Backups and Disaster Recovery
Incremental snapshots enable efficient backup strategies. Automated snapshot schedules back up data regularly, while global snapshot availability enables cross-region disaster recovery.
Log Aggregation and Analytics Workloads
Standard HDD is suitable for log aggregation and data lake storage, where high throughput matters more than IOPS. Combined with Cloud Storage for long-term archiving, Persistent Disk provides cost-effective intermediate storage for analytics pipelines.
Best Practices
Choose the Right Disk Type
Balanced SSD is a good default for most production workloads. Use Standard HDD only for archives or sequential reads. For very high performance requirements, check whether Hyperdisk is available for your machine type before choosing Extreme SSD.
Use Snapshot Schedules
Automate backups with snapshot schedules instead of manual snapshots. Define retention policies matching your compliance requirements. Snapshots are incremental and therefore cost-effective even with frequent creation.
Regional Persistent Disks for Critical Workloads
Use Regional Persistent Disks for databases and other critical services requiring high availability. Synchronous replication across two zones protects against zonal failures but introduces additional latency due to cross-zone replication.
Monitor IOPS and Throughput
Use Cloud Monitoring for disk metrics like read/write IOPS, throughput, and queue depth. High queue depth indicates I/O bottlenecks and can signal a need to switch to a higher tier or to Hyperdisk.
Encrypt Sensitive Data
Use Customer-Managed Encryption Keys (CMEK) for sensitive data to maintain full control over encryption keys. Standard encryption-at-rest is always enabled.
Google Cloud Persistent Disk Comparison
vs. AWS EBS: Both offer multiple performance tiers, automatic replication, and snapshot features. The exact tier structure and pricing models differ between providers and should be compared case by case.
vs. Azure Managed Disks: Azure similarly distinguishes between HDD, Standard SSD, Premium SSD, and Ultra Disk options. Both providers offer comparable availability and backup features.
vs. STACKIT Block Storage: STACKIT offers data sovereignty in German data centers, while Persistent Disk provides global reach with EU regions. STACKIT suits compliance-critical workloads, Persistent Disk suits globally distributed scaling.
Integration with innFactory
As a certified Google Cloud partner, innFactory supports you in selecting the right disk type (Persistent Disk or Hyperdisk), migrating from on-premises storage to GCP, and optimizing storage costs. We help with snapshot strategies, performance tuning, and integration with Kubernetes and Compute Engine.
Contact us for consulting on Google Cloud Persistent Disk and storage architectures.
Available Tiers & Options
Standard (HDD)
- Cost-effective
- Good for sequential workloads
- Up to 64 TiB per disk
- Lower IOPS and throughput
- Not for latency-sensitive applications
Balanced (SSD)
- Balanced price-performance ratio
- Good performance for databases
- Higher IOPS than Standard
- More expensive than Standard HDD
- Lower performance than Extreme SSD or Hyperdisk
Performance (SSD)
- High IOPS and throughput
- Low latency
- Ideal for databases and analytics
- Higher costs
- Overkill for simple workloads
Extreme (SSD)
- High, provisionable IOPS
- Low latency
- For demanding database workloads
- Most expensive Persistent Disk option
- Only usable with certain machine types
Typical Use Cases
Frequently Asked Questions
What is Google Cloud Persistent Disk?
Persistent Disk is a network-based block storage service for Google Compute Engine VMs. The service offers automatic replication within a zone, incremental snapshots for backups, and multiple performance tiers from Standard HDD to Extreme SSD.
Which disk type should I choose?
Standard HDD is suitable for sequential workloads and archives. Balanced SSD offers a good price-performance ratio for databases and applications. Performance SSD suits I/O-intensive workloads, and Extreme SSD suits workloads with very high IOPS requirements. For the highest performance on current machine types, Google now recommends Hyperdisk over Persistent Disk.
How do snapshots work with Persistent Disk?
Snapshots are incremental backups that only store changed blocks since the last snapshot, and can be created even from running VMs. They are globally available and can be restored in other regions for disaster recovery.
Can I move Persistent Disks between VMs?
Yes, Persistent Disks can be detached from one VM and attached to another. In read-only mode, disks can also be attached to multiple VMs simultaneously, enabling flexible workload migration and shared storage scenarios.
How is Persistent Disk billed?
Persistent Disk is billed based on provisioned storage per GB/month, regardless of actual usage. See the official Google Cloud pricing page for regional details.
What is the difference between Persistent Disk and Hyperdisk?
Persistent Disk is the established, network-based block storage option with four fixed tiers, where performance scales with disk size. Hyperdisk is Google's newer generation with IOPS/throughput performance configurable independently of capacity and higher performance limits; some of the newest machine types support only Hyperdisk.
Is Persistent Disk GDPR compliant?
Persistent Disk is available in EU regions. As with all Google Cloud services, GDPR compliance depends on the chosen region, contractual basis, and specific configuration; see Google's compliance documentation for details.
Note: All product information on this page has been compiled with care, but is provided without guarantee and may be outdated or incomplete. Cloud services evolve rapidly — features, pricing, SLAs, and availability change frequently. Authoritative and up-to-date information can only be found on the official product page of Google Cloud (official documentation). This page does not represent an offer by Google Cloud.
