Skip To Content

Strategies for loading data into a geodatabase on Amazon Web Services

As discussed in Strategies for data transfer to AWS, much of your source GIS data will stay in a geodatabase or files you store on-premises, and you will publish services from this data. In some cases, though, the data needs to move into a geodatabase on Amazon Web Services (AWS). To achieve this, you can use the methods described in the following sections.


You can replicate data from an on-premises geodatabase to a geodatabase in your ArcGIS Enterprise on Amazon Web Services deployment.

At 10.6 and previous releases, when you create an ArcGIS Server site using ArcGIS Server Cloud Builder on Amazon Web Services or a CloudFormation template from Esri, a geodatabase named geodata is created with your deployment. Register the geodata geodatabase as a replicated database and create a geodata service from it. You can then create a replica of your local data and replicate it to the geodata service. As edits are made, you can synchronize changes through the geodata service to the geodata geodatabase. See Use a geodata service and a connected replica for more information.

To replicate after 10.6, you'll need to configure the geodatabase and service manually. See Suggestions for configuring geodata services in the ArcGIS Server help for more information.


Similarly, when you create an ArcGIS Server site using Cloud Builder 10.6 or an Esri 10.6 CloudFormation template, a geodatabase is created (egdb) and registered as the ArcGIS Server site's managed database. When you publish a feature or WFS-T service, a copy of the data will be moved to the managed database. When you delete the feature or WFS-T service, the corresponding data is deleted from the managed database.

ArcGIS Enterprise deployments include ArcGIS Data Store. The relational data store configured through ArcGIS Data Store acts as the ArcGIS Server site's managed database. When you publish feature services to the ArcGIS Server site or publish hosted feature layers to the Portal for ArcGIS component of your deployment, data is copied to the relational data store.

When you use an Esri provided CloudFormation template to deploy ArcGIS Enterprise, a tile cache data store is also created. This allows you to publish hosted scene layers to the portal, and the data will be copied to the tile cache data store.

You also have the option to choose to copy data to the ArcGIS Server site when publishing map and feature services.

Load data you moved to the cloud

If necessary, you can also physically move data to the egdb or geodata geodatabases in your deployment on AWS or into geodatabases you create yourself on AWS.

First, you must move the data to Amazon Web Services. Once your source data has been moved to AWS, log in to the instance where ArcGIS Desktop is installed, connect to the source data and the geodatabase on AWS, and load the data into the geodatabase.

From an ArcGIS Desktop client, you can do the following:

You can also transfer databases through backup files or, for SQL Server, the database file itself. You need to move the file to the AWS instance that contained the PostgreSQL or SQL Server database management system included with your ArcGIS Server (enterprise) sites on AWS, connect to the database management system through a SQL client, and restore a dump file from an existing geodatabase in PostgreSQL to your PostgreSQL database cluster or attach a database to SQL Server on AWS that you detached from another SQL Server instance.

For more information on getting files and other source data onto an Amazon Web Services instance, see Strategies for data transfer to Amazon.