Data Science Workshop (DSW) is a cloud-based integrated development environment (IDE) for AI development that supports Notebook, VS Code, and Terminal environments. DSW also provides built-in images for mainstream AI frameworks such as PyTorch and TensorFlow, supports a wide range of heterogeneous compute resources, and can mount datasets from OSS, NAS, and CPFS to establish an efficient development workflow.
Product overview
The following figures show the development environment of DSW.
New UI
Classic UI
Benefits
Flexible and easy to use: Integrates multiple development environments and supports images for open-source frameworks such as PyTorch and TensorFlow. DSW also provides heterogeneous compute resources, including public resources and dedicated resources such as general-purpose compute resources or Lingjun intelligent computing resources.
End-to-end service: DSW provides tools such as PAI-DLC for distributed training and PAI-EAS for model online service. This creates an end-to-end AI development pipeline, from data processing, development, and debugging to model training and deployment.
Granular management: You can configure lifecycle management settings such as scheduled shutdown and idle shutdown to save costs. The workspace feature supports global resource allocation and reclamation.
Scenario-based examples: The Notebook Gallery provides tutorials and examples for cutting-edge areas such as LLMs and AIGC, so you can get started fast or build on them.
Core features
Create and manage
Create a DSW instance: When you create a DSW instance, you can select an instance resource type, mount datasets, and specify a custom image.
Access and manage DSW from the console: You can use the comprehensive features of DSW in the console and perform common operations, such as stopping, releasing, and reconfiguring instances.
Use an instance RAM role: Associate a RAM role to access other cloud resources from within the instance by using temporary STS credentials. This removes the need to configure a long-term AccessKey and reduces the risk of key exposure.
Model development environment
Manage third-party libraries: Manage and install third-party Python libraries or software.
Visualize training with TensorBoard: Use the TensorBoard plug-in to visualize metrics and information during model training.
Deploy a model as an online service: When you need to call a model from other applications and require elastic scaling, version control, and resource monitoring, use PAI-EAS to deploy it as a model online service.
Manage secondary containers with DockerBoard and Use Docker in DSW: Create and manage secondary containers within a DSW instance.
Data access and mounting
Mount datasets/OSS/NAS/CPFS: You can mount a dataset or an OSS, NAS, or CPFS path to expand the storage space of an instance, persist data, and read data files.
Read and write data in OSS: You can read data from and write data to files in OSS in a DSW instance by using an API or SDK.
Upload and download files: You can transfer data and models between your local machine and an instance.
Network configuration
Connect remotely via SSH: An SSH remote connection provides a local development experience while leveraging the powerful computing capabilities of DSW.
Improve public network access speed with a NAT Gateway: You can create a NAT Gateway for the instance's VPC and bind an EIP to improve the instance's network upload and download speeds.
Access services in an instance from the internet: You can access the services running in an instance from a VPC or the internet. This is useful for model testing and verification.
Pull models or container images from overseas: You can configure Global Accelerator (GA) for DSW to accelerate pulling container images from regions outside the Chinese mainland, such as docker.io images, or models, such as huggingface.co models.
Billing
Compute instances
You can select public resources or dedicated resources, such as general-purpose compute resources or Lingjun intelligent computing resources, as the instance type. Different billing methods apply to different instance types.
Instance type | Billing method | Billable item | Billing rules | Stop billing |
public resource | pay-as-you-go | DSW instance runtime. | You are charged based on the runtime of DSW instances created with public resources. Important Note on billing: DSW instances are billed by the minute, and bills are generated hourly. Due to data aggregation and processing, bill generation may be delayed by 2 to 3 hours. The final bill prevails. | Stop or delete the DSW instance. Important You can stop an instance manually or configure a scheduled shutdown. For more information, see Manage DSW instances. |
dedicated resource (general-purpose compute resource or Lingjun intelligent computing resource) | subscription | The number of purchased nodes and the subscription duration. | After purchasing dedicated resources, you are charged based on the number of nodes and the subscription duration. For more information, see Billing of AI computing resources. | Unsubscribe from the resource. |
System disk
Billing method | Billable item | Billing rules | Stop billing |
pay-as-you-go | System disk capacity and usage duration. | Each instance type and specification includes a free quota. You can expand capacity. You are billed for any expanded capacity based on its size and usage duration. | Delete the DSW instance. |
For more information about billing, see Billing of Data Science Workshop (DSW). To view your bill details, see View your bill details.
Quick start
If you are new to DSW, read Quick Start for DSW first. It uses MNIST handwritten digit recognition as an example to help you get started with DSW.
Get help
If you encounter issues such as instance startup or stop failures, billing questions, free trial expiration, remote connection failures, low download speeds, or issues with accessing DSW over the internet, see FAQ about DSW.