This page provides a detailed overview of the Google Cloud NetApp Volumes storage pools feature.
About storage pools
Storage pools act as a container for volumes. All volumes in a storage pool share the same location, service level, Virtual Private Cloud network, Active Directory policy, and customer-managed encryption key (CMEK) policy. You can assign the capacity of the pool to volumes within the pool. Billing is based on the location, service level, and capacity allocated to a pool independent of consumption at the volume level.
Supported locations
All volumes in a storage pool live in the same location as the pool. For more information on supported locations, see NetApp Volumes supported locations.
Service levels
The service level is a property of the storage pool. Service levels have capacity-dependent throughput limits. NetApp Volumes offers four service levels: Flex, Standard, Premium, and Extreme. See NetApp Volumes service levels table to learn more.
The following table provides the throughput of allocated capacity for each service level and what each service level is best used for:
| Service level | Throughput | Workload types |
|---|---|---|
| Flex | 16 KiBps per GiB of allocated pool capacity | General purpose storage |
| Standard | 16 KiBps per GiB of allocated volume capacity | General purpose storage |
| Premium | 64 KiBps per GiB of allocated volume capacity | Databases and applications |
| Extreme | 128 KiBps per GiB of allocated volume capacity | High-performance applications |
You can reassign the Premium service level volume to an Extreme service level storage pool and the other way around. This allows instant adjustments to changing performance requirements.
Availability
Pools in different service levels have different availability, see NetApp Volumes service level agreement. Pools of Flex service level lets you to choose between zonal and regional availability.
A regional pool maintain two synchronous replicas of your data in two different zones. When creating a regional pool, you can choose the active zone and a replica zone. All volumes of your pool are available from the active zone, while the replica zone acts as a standby. In case the active zone fails or you are triggering a manual zone switch, all volume access moves to the replica zone, which then becomes the new active zone. The previous active zone becomes the new replica zone. This zone switch can cause a pause of file service activity. After the zone switch is completed, the file services starts again.
When you place the active zone into the same zone as your workload VMs optimizes I/O latency and helps you achieve more performance.
Network
A pool defines which Virtual Private Cloud (VPC) its volumes connect to. The volumes are accessible only for clients that are on the same VPC or those connected to it with a VPN.
Active Directory settings
Some network-attached storage (NAS) protocols or protocol variations require access to an external directory service. NetApp Volumes supports Active Directory as the only directory service. A single pool can only attach one Active Directory policy. All volumes in the pool use the same policy.
Customer-managed encryption key settings
NetApp Volumes encrypts the data of all volumes at rest using a Google-owned and Google-managed key, however, you can choose to encrypt the volumes using a customer-managed encryption key (CMEK). You can attach a CMEK policy during storage pool creation; all of the volumes created in that pool use the policy.
Capacity
NetApp Volumes lets you add or remove capacity to your volume as needed to adjust for any application changes. Capacity is allocated to storage pools and volumes, and can be modified. The following table shows how you can allocate capacities to NetApp Volumes storage pools and volumes for each service level.
| Features | Service level type | |||
|---|---|---|---|---|
| Flex | Standard | Premium | Extreme | |
| Storage pools | ||||
| Capacity | 1 to 200 TiB | 2 to 200 TiB | 2 TiB to 10 PiB | 2 TiB to 10 PiB |
| Granular resizing | Yes, the storage pool can increase in 1 GiB increments | Yes, the storage pool can increase and decrease in 1 GiB increments | Yes, the storage pool can increase and decrease in 1 GiB increments | Yes, the storage pool can increase and decrease in 1 GiB increments |
| Volumes | ||||
| Capacity | 1 GiB to 102,400 GiB | 100 GiB to 102,400 GiB | 100 GiB to 1 PiB | 100 GiB to 1 PiB |
| Granular resizing | Yes, the volume can increase and decrease in 1 GiB increments | |||
Auto-tiering (Preview)
Auto-tiering reduces the overall cost of volume usage. It can be enabled at the per volume level. Users who have large amounts of inactive data can reduce their overall storage cost with auto-tiering.
Auto-tiering needs to be allowed on a storage pool before auto-tiering can be enabled for any of the volumes in the pool. Once you allow auto-tiering for the pool, the selection is permanent. Allowing auto-tiering on the pool does not change existing volumes, nor does it affect how new volumes are created by default. Instead, you have to explicitly enable auto-tiering for individual volumes by updating existing volumes or when creating a new volume.
Volumes
You can assign volumes to storage pools, and also move a volume within a pool to a different pool with the same settings (location, network, Active Directory policy, and CMEK policy) as long as the service level allows and the target pool has enough spare capacity.
Before you can start to create volumes, you need to provision a storage pool to host the volumes.