Compute engines - Dataphin - Alibaba Cloud Documentation Center

Once the tenant's computing settings, you can add the appropriate compute engines to Dataphin. These engines can be linked to Dataphin projects to provide the necessary computational power for processing tasks.

Feature overview

You can add compute engines as per your business requirements. Below are the descriptions and instructions for setting up each type of compute engine:

Computing settings	Compute engines
MaxCompute	For tenants using MaxCompute as their computing setting, the following compute engines can be created: Offline compute engines MaxCompute Compute Engines: Projects linked to MaxCompute compute engines enable standard visual modeling, ad hoc queries, MAXC SQL tasks, general scripting, and additional features. For instructions on creating MaxCompute compute engines, see Create MaxCompute Compute Engines. Hologres compute engines support ad hoc queries, Hologres SQL compute tasks, and additional features in attached projects. For instructions on creating Hologres compute engines, see Create Hologres compute engines. Machine Learning PAI: Dataphin seamlessly integrates with Platform for AI, offering essential algorithm scheduling features of Machine Learning PAI. When setting up a workspace for visual modeling on the machine learning platform, ensure to select a MaxCompute-based computing resource group. For more information, see AI computing resources. Hologres Accelerated Compute Engines: Dataphin allows the addition of Hologres accelerated compute engines to projects that are already utilizing MaxCompute compute engines, enhancing the query performance of business data. For instructions on creating Hologres accelerated engines, see Create Hologres Compute Engines. Real-time compute engines Blink Compute Engines: Projects linked to Alibaba Blink can leverage Blink tasks. To establish Blink compute engines, refer to Create Blink Compute Engines. Ververica Flink compute engines: Projects attached to Ververica Flink compute engines support Flink tasks. For creating Blink compute engines, see Create Ververica Flink compute engines.
Hadoop	Tenants with Hadoop as their computing setting can create the following compute engines: Offline compute engines Hadoop compute engines: These engines support standard modeling, ad hoc queries, Hive SQL tasks, and general scripts. For instructions on creating Hadoop compute engines, see Create Hadoop compute engines. Real-time compute engines Flink compute engines: These engines are optimized for Flink tasks. For setup instructions, see Create Flink compute engines.
AnalyticDB for PostgreSQL	Tenants with AnalyticDB for PostgreSQL as their computing setting can create the following compute engines: Offline compute engines AnalyticDB for PostgreSQL compute engines: These engines support ADB SQL tasks, ad hoc queries, and general scripts. For more details, see Create AnalyticDB for PostgreSQL compute engines. Real-time compute engines Ververica Flink compute engines: These engines are designed for Flink tasks. For creation instructions, see Create Ververica Flink compute engines. Flink compute engines: Optimized for Flink tasks, these engines can be created by following the steps in Create Flink compute engines.
Transwarp TDH 6.x	Tenants with Transwarp TDH 6.x as their computing setting can create the following compute engines: Offline compute engines TDH Inceptor compute engines: These engines support Inceptor SQL tasks. For setup instructions, see Create TDH Inceptor compute engines. Real-time compute engines Flink compute engines: Optimized for Flink tasks, these engines can be set up as detailed in Create Flink compute engines.
Transwarp ArgoDB	Tenants with Transwarp TDH 6.x as their computing setting can create the following compute engines: Offline compute engines ArgoDB compute engines: These engines support ad hoc queries and ArgoDB SQL compute tasks. For instructions on creating ArgoDB compute engines, see Create ArgoDB compute engines. Real-time compute engines Not supported.