ArcGIS Enterprise supports configuring web services that reference user data, as an alternative to copying the data to ArcGIS Enterprise during publishing. When ArcGIS Enterprise is running in the cloud, additional data storage options become available to support referenced web layers. These include cloud storage, managed database services, and cloud data warehouses.
The options available will depend on the type of web service you are publishing as well as the cloud ArcGIS Enterprise is running in.
Cloud storage
For cloud deployments, you can use cloud storage to store the data and caches for referenced web layers. The types of layers and the cloud storage services that are supported vary by workflow, but you can read more about them by following the links below.
Cloud storage locations can be registered as data stores with ArcGIS Enterprise and used in the following ways:
- Store prerendered cached tile images for map services or image services configured to draw from a cache. Using cloud data stores as cache directories is an alternative to storing large caches on your file system.
- Store prerendered caches for tile, vector tile, 3D tiles, or scene layers. Caches will be created and placed in the registered cloud storage location, with referenced services being published directly from the data store item in ArcGIS Enterprise. This approach saves time while publishing large datasets and avoids using resources on the hosting server, relative to the alternative of copying all data while publishing from ArcGIS Pro.
- Store raster and image files that can be used to publish referenced imagery layers. This is supported when ArcGIS Enterprise is configured for image hosting or raster analysis.
Learn how to add a cloud storage data store item in the portal or register a cloud storage location using ArcGIS Server Manager.
Database services
Cloud vendors provide Databases as a Service (DBaaS) that allow you to use database systems without the need to install, upgrade, or maintain the underlying software and hardware. These services also provide significant benefits in terms of manageability and scalability. ArcGIS supports various DBaaS solutions, including support for storing enterprise geodatabases in them. This can be an alternative to installing a relational database management system (RDBMS) such as Microsoft SQL Server, Oracle, or PostgreSQL on a virtual machine (VM) in the cloud.
For the list of supported database service offerings, see the requirements for using ArcGIS with databases in the cloud.
Cloud data warehouses
Like cloud storage and database services, cloud data warehouses are scalable, resilient, and performant solutions for storage that are available in public clouds as managed services. Cloud data warehouses are optimized for analytics on large volumes of data and are increasingly used as a centralized repository for data storage. ArcGIS Enterprise supports integrating data from cloud data warehouses for visualization and analysis.
See Requirements for using ArcGIS with databases in the cloud for more information about supported cloud data warehouses for use with ArcGIS Enterprise and ArcGIS Pro.
Latency considerations for cloud services integration
Using cloud services for user data creates the potential for latency that can negatively impact performance. For example, registering a cloud data warehouse running in a cloud provider's Eastern US region with an ArcGIS Server site running on-premises in an Australian data center will require requests for data to travel a long distance over the network. Services that are published using that data will be slow to respond to requests.
To minimize latency, it is recommended that you co-locate components in the same region of the same cloud provider. Co-location reduces the distance information needs to travel across the network. For example, if you are using Amazon S3 in the AWS af-south-1 region to store imagery for your referenced imagery layers, all of the other components of ArcGIS Enterprise should also be running in the AWS af-south-1 region. Clients that publish services to ArcGIS Enterprise, such as ArcGIS Pro, should also be co-located with ArcGIS Enterprise.
Note:
Esri does not formally support connecting to cloud services for referenced data if ArcGIS Enterprise software components are not running in the same region of the same cloud.