All Products
Search
Document Center

Platform For AI:AI computing resources

Last Updated:Feb 27, 2026

Before using Platform for AI (PAI), activate the service and purchase computing resources for AI development and training. Cloud-native resources require purchasing resources, creating resource quotas, and associating quotas with workspaces. Big data engine resources only need to be purchased and associated with workspaces. This topic describes both cloud-native and big data engine resource types.

Resource types

  • Cloud-native resources

    • Lingjun resources

      Lingjun resources deliver large-scale deep learning and integrated intelligent computing capabilities. Through seamless hardware-software integration, Lingjun provides high-performance heterogeneous computing power essential for AI workloads. These resources offer high performance, efficiency, and resource utilization for AI development, training, and inference in PAI.

    • General computing resources

      General computing resources include Elastic Compute Service (ECS), Elastic Container Instance, and Elastic GPU Service (EGS), providing flexible, stable, and high-performance environments for deep learning model training. After PAI activation, the system automatically creates a public quota for general computing resources. Associate this quota with a workspace to begin using the resources.

  • Big data engine resources

    • MaxCompute resources

      MaxCompute is an enterprise-level SaaS cloud data warehouse designed for data analytics scenarios. It provides fully managed online data warehousing with fast performance, eliminating traditional platform constraints on resource scalability and elasticity. MaxCompute minimizes operations costs while enabling efficient processing and analysis of large-scale data at low cost. For more information about MaxCompute resources, see What is MaxCompute?

    • Fully managed Flink resources

      Realtime Compute for Apache Flink is Alibaba Cloud's comprehensive real-time big data analytics platform built on Apache Flink. It supports end-to-end data analytics with subsecond latency. For more information, see What is Realtime Compute for Apache Flink?

Guidance

Log on to the PAI console. In the left-side navigation pane, choose AI Computing Resources > Resource Pool. On the Resource Pool page, create resource groups and purchase computing resources.

Log on to the PAI console. In the left-side navigation pane, choose AI Computing Resources > Resource Quota. On the Resource Quota page, create resource quotas for different resource types. Associate these quotas with workspaces to perform AI development and training.

  • Cloud-native resource quotas

    After purchasing cloud-native resources on the Resource Pool page, allocate computing resources from one or more resource groups to resource quotas. Create hierarchical child quotas to build a quota tree that enhances queuing and scheduling performance.

  • Big data resource quotas

    • For more information about how to activate MaxCompute, purchase MaxCompute resources, and create and use a resource quota for MaxCompute resources, see MaxCompute resources.

    • For more information about how to activate Realtime Compute for Apache Flink, purchase fully managed Flink resources, and create and use a resource quota for the resources, see Fully managed Flink resources.