Skip to main content

S3 configuration overview

Contributors netapp-ahibbard netapp-dbagwell netapp-aoife netapp-forry netapp-thomi johnlantz netapp-aherbin
PDFs

Beginning with ONTAP 9.8, you can enable an ONTAP Simple Storage Service (S3) object storage server in an ONTAP cluster, using familiar manageability tools such as ONTAP System Manager to rapidly provision high-performance object storage for development and operations in ONTAP and taking advantage of ONTAP's storage efficiencies and security.

The primary use cases for S3 in ONTAP are:

  • Using FabricPool to tier inactive data to a bucket in ONTAP, allowing for ONTAP-to-ONTAP tiering. Tiering to a bucket within the local cluster or tiering to a bucket on a remote cluster are both supported. Tiering to ONTAP S3 lets you use less expensive ONTAP systems for inactive data and save money on new flash capacity without the need for additional FabricPool licenses or new technologies to manage.

  • Beginning in ONTAP 9.12.1, the S3 protocol can also be enabled in multiprotocol NAS volumes that have been preconfigured to use NAS protocols. When the S3 protocol is enabled in multiprotocol NAS volumes, client applications can read and write data using S3, NFS, and SMB, which opens up a variety of additional use cases. One of the most common use cases is NAS clients writing data to a volume and S3 clients reading the same data and performing specialized tasks such as analytics, business intelligence, machine learning, and optical character recognition.

Note ONTAP S3 is appropriate if you want to enable S3 capabilities on existing ONTAP clusters without additional hardware and management. NetApp StorageGRID is NetApp's flagship solution for object storage. StorageGRID is recommended for native S3 applications that need to take advantage of the full range of S3 actions, advanced ILM capabilities, or capacities not achievable in ONTAP-based systems. For more information, see the StorageGRID documentation.

S3 configuration with System Manager and the ONTAP CLI

You can configure and manage ONTAP S3 with System Manager and the ONTAP CLI. When you enable S3 and create buckets using System Manager, ONTAP selects best-practice defaults for simplified configuration. If you need to specify configuration parameters, you might want to use the ONTAP CLI. If you configure the S3 server and buckets from the CLI, you can still manage them with System Manager if desired, or vice-versa.

When you create an S3 bucket using System Manager, ONTAP configures a default performance service level that is the highest available on your system. For example, on an AFF system, the default setting would be Extreme. Performance service levels are predefined adaptive Quality of Service (QoS) policy groups. Instead of one of the default service levels, you can specify a custom QoS policy group or no policy group.

Predefined adaptive QoS policy groups are:

  • Extreme: Used for applications that expect the lowest latency and highest performance.

  • Performance: Used for applications with modest performance needs and latency.

  • Value: Used for applications for which throughput and capacity are more important than latency.

  • Custom: Specify a custom QoS policy or no QoS policy.

If you select Use for tiering, no performance service levels are selected, and the system tries to select low-cost media with optimal performance for the tiered data.

See also: Use adaptive QoS policy groups.

ONTAP tries to provision this bucket on local tiers that have the most appropriate disks, satisfying the chosen service level. However, if you need to specify which disks to include in the bucket, consider configuring S3 object storage from the CLI by specifying the local tiers (aggregate). If you configure the S3 server from the CLI, you can still manage it with System Manager if desired.

If you want the ability to specify which aggregates are used for buckets, you can only do so using the CLI.

Configuring S3 buckets on Cloud Volumes ONTAP

If you want to serve buckets from Cloud Volumes ONTAP, it is strongly recommended that you manually select the underlying aggregates to ensure that they are using one node only. Using aggregates from both nodes can impact performance, because the nodes will be in geographically separated availability zones and hence susceptible to latency issues. Therefore, in Cloud Volumes ONTAP environments, you should configure S3 buckets from the CLI.

Otherwise, S3 servers on Cloud Volumes ONTAP are configured and maintained the same in Cloud Volumes ONTAP as in on-premises environments.