Create resource groups and purchase Lingjun intelligent computing resources - Platform For AI

Overview

Lingjun intelligent computing resources feature RDMA high-speed networking, high-performance communication libraries and acceleration software, and GPU virtualization for demanding AI workloads.

PAI provides fully managed Lingjun resources that you can purchase and manage entirely in the PAI console. If you have purchased standalone Lingjun hardware, you can add it to PAI as semi-managed resources for training jobs.

Limitations

Supported regions

Available only in China (Ulanqab), Singapore, China (Shenzhen), China (Beijing), China (Shanghai), and China (Hangzhou).
Supported users

Available in subscription and pay-as-you-go billing for whitelisted users only. Contact your business manager for access.
Supported task types

Support TensorFlow, PyTorch, ElasticBatch, XGBoost, OneFlow, MPIJob, Slurm, and Ray training jobs only.

Account and permissions

Alibaba Cloud account (Recommended): Perform all operations without additional permissions.
RAM user: Your Alibaba Cloud account must grant permissions to manage resource pools or grant AliyunPAIFullAccess permission to the RAM user.

Important
AliyunPAIFullAccess grants full control over all PAI resources and features. Grant this permission with caution. We recommend using your Alibaba Cloud account for these operations.

Dependencies

Lingjun intelligent computing resources require other cloud products. Review these dependencies and prepare necessary resources before you begin.

(Required) VPC

Bind Lingjun resources to a VPC in the same region and configure a vSwitch and security group. This ensures network connectivity between Lingjun resources and other cloud products.

(Optional) Internet NAT Gateway and EIP

To access the internet (for example, to use custom images from public repositories), enable and configure the SNAT feature of an Internet NAT Gateway and bind an EIP. This enables Lingjun resources to access the internet.

For more information about how to configure the SNAT feature, see Use the SNAT feature of an Internet NAT Gateway to access the internet.

(Optional) OSS, NAS, or CPFS for AI

DLC jobs on Lingjun resources require datasets. PAI supports datasets from OSS, NAS, and CPFS for AI. Prepare the necessary storage resources and create datasets. For more information, see Preparations: Prepare datasets.

Procedure

Lingjun resources are available with subscription or pay-as-you-go billing. After creating a resource group, you can purchase resources. For more information about billing, see Billing of AI computing resources.

Note

Lingjun divides underlying computing resources into High-speed Interconnect Zones (HZs). Resources with the same HZ ID communicate over a high-speed network. Customize high-speed network affinity when purchasing, allocating, and using resources.

Create a resource group

Log on to the PAI console and navigate to AI Computing Resources > resource pool.
On the Lingjun Intelligent Computing resources tab, click Create Resource Group.

In the Create Resource Group dialog box, configure the parameters and click OK.

Parameter	Description
Type	Select Dedicated Resource Group.
Resource Group Name	Enter a name that follows the on-screen naming conventions.

Purchase subscription resources

On the Intelligent Computing LINGJUN Resources tab, find the resource group. In the Actions column, click Create Order > Create Subscription Order.

Alternatively, click the resource group name. In the upper-right corner of the details page, click Create Order > Create Subscription Order.
Resource group ID and region are automatically selected. Select Node Specification, Quantity, and Duration, then click Buy Now.

Common issues and solutions:
- Order does not contain information for the current resource group.
  - Cause: Selected region does not match the resource group's region.
  - Solution: Switch to the default region of the resource group.
- Specified instance type is out of stock in zone.
  - Cause: Selected node specification is temporarily out of stock in the current region.
  - Solution: Select a different node specification.
- Current kind of instance is temporarily not supported. Choose another kind of ECS to purchase.
  - Cause: Selected node specification is not supported for purchase in the current region.
  - Solution: Select a different node specification.
After payment, purchased subscription resources appear on the order list page.The system splits each purchase into separate orders per node, allowing individual management such as renewing or unsubscribing.

Purchase pay-as-you-go resources

Note

Currently available to whitelisted users only. Contact your business manager for access.

On the Lingjun Intelligent Computing resources tab, find the resource group. In the Actions column, click Create Order > Create Pay-As-You-Go Order.

Alternatively, click the resource group name. In the upper-right corner of the details page, click Create Order > Create Pay-As-You-Go Order.
Region, Resource Type, and Resource Group are pre-selected. Select Resource Specification, set Quantity, then click OK.
After purchase, filter the order list by billing method to view subscription or pay-as-you-go orders.

Important
To avoid unnecessary charges, stop unneeded pay-as-you-go Lingjun resources. On the Order Information page, click Stopped in the Actions column. Stopping a node causes running jobs on that node to fail. Proceed with caution.

Next steps

After creating a resource group and purchasing computing resources:

View the resource group's basic information and manage your purchased resources on the resource group page. For more information, see Manage resources.
Configure a resource quota to allocate these resources for specific training jobs. For more information, see Create a resource quota.