Function overview
You can add compute sources as needed. The following table describes each compute source and provides instructions on how to create them.
Compute setting | Compute source |
MaxCompute | If a tenant's compute setting is MaxCompute, you can create the following compute sources: Offline compute source MaxCompute compute source: Projects attached to a MaxCompute compute source support features such as standard modeling, ad hoc queries, MAXC SQL nodes, and generic scripts. To create a MaxCompute compute source, see Create a MaxCompute compute source. Hologres compute source: Projects attached to a Hologres compute source support features such as ad hoc queries and Hologres SQL compute nodes. To create a Hologres compute source, see Create a Hologres compute source. Platform for AI (PAI): Dataphin integrates with Platform for AI (PAI) to provide basic algorithm scheduling capabilities from PAI. In PAI, when you create a workspace for visualization modeling, select a MaxCompute-based compute resource group for the resource. For more information, see Overview of AI compute resources. Hologres accelerated compute source: Add a Hologres accelerated compute source to projects already attached to a MaxCompute compute source to speed up data queries. To create a Hologres acceleration engine, see Create a Hologres compute source.
Real-time compute source Blink compute source: Projects attached to an Alibaba Blink compute source support Blink nodes. To create a Blink compute source, see Create a Blink compute source. Ververica Flink compute source: Projects attached to a Ververica Flink compute source support Flink nodes. To create a Ververica Flink compute source, see Create a Ververica Flink compute source.
|