There are two basic types of databases in the cloud: databases that are installed on virtual machines in the cloud, and databases that are available as database-as-a-service offerings.
When you use a database installed on a virtual machine in the cloud, you have access to the virtual machine where the database is installed and are usually responsible for maintaining it. This is generally true whether you install the database on the virtual machine yourself or you use an image to launch an instance that includes an installation of the database. You are also responsible for most maintenance tasks on these types of databases.
This is different from database-as-a-service offerings. For these types of databases, you do not have access to the machine that serves up the database, and you do not have to perform maintenance tasks, such as applying security patches, on the database or operating system.
Connections from ArcGIS software to databases in the cloud must originate from machines in the same cloud. For best performance, connections should originate from machines in the same cloud region. In the majority of cases, connection and query performance is poor when you connect from ArcGIS clients installed on-premises to databases in the cloud or make connections between clients and databases in different cloud regions, and connections may time out.
Support for databases installed in the cloud
To determine if a cloud platform is supported for use with ArcGIS, you need to know the specifications of the cloud virtual machines. A cloud platform is supported for use with a database that ArcGIS supports as long as the virtual machines provided on that cloud platform meet the operating system and system specification requirements for use with the ArcGIS clients and database you want to install.
Support for these virtual machines is predicated on the assumption that the virtual machines behave the same as a physical machine available on-premises that has the same specifications. In most cases, Esri Support attempts to replicate issues using on-premises machines with the same specifications as the cloud virtual machine you are using.
Supported cloud-based database services
Esri provides tooling to deploy ArcGIS Enterprise software, ArcGIS Desktop software, and many supported database-as-a-service offerings on Amazon Web Services (AWS) and Microsoft Azure. When deploying on these two cloud platforms, it's recommended that you use the specialized tooling and Esri images to simplify deployment and technical support. For cloud platforms for which Esri does not provide specialized deployment tooling, support is limited to troubleshooting ArcGIS software-specific issues that can be replicated on-premises.
The following database-as-a-service offerings are supported for use with ArcGIS. The links below take you to external documentation provided by cloud vendors.
- Amazon Aurora PostgreSQL
- Amazon Relational Database Service (RDS) for Oracle (Support begins with 10.8.1 and ArcGIS Pro 2.6.)
- Amazon RDS for PostgreSQL
- Amazon RDS for SQL Server
- Microsoft Azure Database for PostgreSQL
- Microsoft Azure SQL Database or Microsoft Azure SQL Managed Instance*
- SAP HANA Cloud (Support begins with 10.8.1 and ArcGIS Pro 2.6.)
*Changes Microsoft Azure has made for SQL Managed Instance may prevent deployment using ArcGIS Enterprise Cloud Builder for Microsoft Azure. If you experience issues, use Microsoft Azure portal to create the instance.
Geoprocessing tools are supported on equivalent database-as-a-service offerings unless otherwise stated in the geoprocessing tool usage information. For example, if tool usage indicates the tool is supported with PostgreSQL, it works with Amazon RDS for PostgreSQL, Aurora PostgreSQL, and Azure Database for PostgreSQL unless the tool's usage statements indicate these are not supported.
Limitations
Because you cannot load the ST_Geometry library to a database-as-a-service offering, only the PostGIS spatial types are supported when using Amazon Aurora PostgreSQL, Amazon RDS for PostgreSQL, and Microsoft Azure Database for PostgreSQL and only the SDO_Geometry spatial type is supported when using Amazon RDS for Oracle.
The following functions are not present in geodatabases in Amazon RDS for Oracle: GDB_UTIL.get_extent(), GDB_UTIL.geometry_type, and GDB_UTIL.spatial_ref_info().
Amazon RDS for Oracle is not currently included in Esri cloud deployment tools; use AWS tools instead.
If you do not use or cannot use the Esri cloud deployment tools to create the database, and you want to create a geodatabase in any of the supported database-as-a-service offerings listed above, you must use the Enable Enterprise Geodatabase geoprocessing tool to create a geodatabase. You cannot use the Create Enterprise Geodatabase tool.
Operating system authenticated connections are not supported from ArcGIS clients to database-as-a-service offerings.
Software required to connect to a database
If you use an Amazon Machine Image (AMI) provided by Esri or an Azure image provided by Esri to deploy ArcGIS Enterprise and ArcGIS Desktop software, the required database clients are already installed that allow you to connect to supported database-as-a-service offerings on AWS and Azure respectively.
If you install an ArcGIS client on a virtual machine in the cloud or connect to a cloud-based database service, you may need to install database management system client files on the ArcGIS client machine. These client files are available from their respective database vendors. Follow instructions from the vendors to install and configure these files, and read the connection information in ArcGIS help for additional configuration needed to connect from ArcGIS clients.
To connect from ArcGIS to SAP HANA Cloud, install SAP HANA client 2.4 revision patch 186 (2.04.186). Be sure to install your ArcGIS client on a machine (physical or virtual) that is located as close as possible to the location where you've deployed SAP HANA Cloud.