How do I maintain consistent operations across multiple AWS Regions in Amazon Quick?

Resolution

Understand regional architecture in Amazon Quick

You can only have one Amazon Quick subscription per AWS account.

When you first subscribe to Amazon Quick, choose a local Region. The local Region stores your user identities, namespaces, and SPICE capacity. Dashboards, analyses, and datasets are stored in the Region where you created them.

Regions separately allocate the SPICE capacity, and the default SPICE capacity is allocated only to your local Region. Other Regions don't have SPICE capacity unless you separately purchase capacity. Note that you can't transfer SPICE capacity between Regions. For more information, see Configure SPICE memory capacity.

User identities are tied to the identity Region. If the identity Region becomes unavailable, then users can't authenticate to Amazon Quick in a Region until the identity Region recovers.

For more information, see AWS Regions, websites, IP address ranges, and endpoints.

Check feature availability by Region

Before you plan your deployment, verify feature availability.

For a list of Regions that Amazon Quick is available, see Supported AWS Regions for Amazon Quick. For a list that Amazon Q is available in Amazon Quick, see Supported AWS Regions for Amazon Q in Quick.

If a feature isn't available in your current Region, then subscribe to Amazon Quick in a Region that supports the feature. Or, you can use cross-region data access.

Resolve cross-region data source connectivity issues

If your data source is in a different Region than your Amazon Quick subscription, then use one of the following solutions for your service.

For Athena

To query data stored in a different Region, create an AWS Glue Data Catalog and Amazon Athena workgroup in the same Region as your Amazon Quick subscription. Then, create a table that points to the cross-region Amazon Simple Storage Service (Amazon S3) bucket location. Athena supports queries for Amazon S3 data across Regions, and then the table in your local Athena can reference S3 data in another Region.

Note: Cross-region queries incur Amazon S3 data transfer charges. Athena federated queries across Regions aren't supported.

For Amazon RDS or Amazon Redshift

To access cross-region data, configure virtual private cloud (VPC) peering between the Amazon Quick Region and the data source Region. Then, set up a VPC connection on Amazon QuickSight to directly access the database.

For more information, see Amazon QuickSight deployment models for cross-account and cross-Region access to Amazon Redshift and Amazon Relational Database Service (Amazon RDS).

Note: If you use cross-region data access pattern, then there are additional network latency and data transfer costs between Regions. Cross-region queries might experience higher latency compared to same-region queries. To avoid repeated cross-region queries, import frequently used datasets into SPICE.

Resolve identity Region mismatch errors

If you attempt user management operations in a Region that isn't your identity Region, then you might receive the following error message:

"Operation is being called from endpoint aa-example-1, but your identity Region is aa-example-2. Please use the endpoint aa-example-2."

To resolve this error, run the DescribeAccountSubscription API operation to identify your identity Region. Then, use the API endpoint for your identity Region to perform user and namespace management operations.

Note: You can't change your identity Region after your initial subscription. If you need to change your identity Region, then you must unsubscribe from Amazon Quick and re-subscribe in a different Region. If you unsubscribe from Amazon Quick, then your assets and users in that Region are deleted.

Migrate assets across Regions

To move dashboards, analyses, datasets, and other assets between Regions, use the Amazon Quick asset bundle export and import API operations. Complete the following steps:

1. To export assets from the source Region, run the StartAssetBundleExportJob API operation.
2. Download the exported bundle.
3. To import the bundle into the target Region, run the StartAssetBundleImportJob API operation.
   Note: Asset bundle API operations migrate only the asset definitions but don't migrate the data that's contained within the assets. You must separately configure data sources, VPC connections, and user permissions in the target Region.

Plan for disaster recovery across Regions

To implement a disaster recovery strategy, regularly back up your Amazon Quick assets to another Region. Then, maintain infrastructure-as-code templates for your Amazon Quick resources to support rapid redeployment. For data layer resilience, make sure that your underlying data sources have their own cross-region replication strategies in place.

For more information, see Resilience in Amazon Quick. I manage the AWS regional differences in Amazon Quick to maintain consistent feature availability across multiple Regions?