Application scenarios

Microservices architecture usage

Leverage Baidu AI Cloud Container Engine (CCE) to break down service modules into smaller units, allowing independent deployment and scaling, simplifying management for users.

Each container can operate with entirely different environment images. Upon initialization, a container creates a unique microservices host node with a distinct network IP. Service discovery between front-end and back-end layers is seamlessly handled via environment variable injection, configuration file mounting, and other methods. Additionally, cloud services on the platform can integrate directly with custom microservices to create robust service clusters.

Migrating or replicating clusters with multiple services is straightforward—select the appropriate Docker service image, configure mutual access addresses, and quickly establish a new identical cluster.

Output of AI technology

It delivers optimal deep learning performance on cloud containers. The Cloud Container Engine offers comprehensive, reliable, and highly available Kubernetes cluster services, combined with large-scale computing, storage, and GPU, NPU, and FPGA support. It enables differentiated AI outputs, such as Baidu's autonomous vehicles, addresses enterprise requirements for Baidu AI adoption, and supports scenarios like rapid deployment, resource sharing and isolation, automated task scheduling, fault recovery, and standardized workflows from model training to deployment.

Improve resource utilization and reduce costs

In cases where users need to run one-off tasks like specific calculations, traditional service models require continuously running servers, leading to resource wastage. By using CCE, containers can be quickly created and destroyed to meet true pay-as-you-go needs. Simply prepare a Docker image for the one-time task, create a container in real-time to execute it, and upon task completion, the container automatically shuts down to free resources.

Continuous integration/continuous delivery

It offers an exceptional DevOps environment with continuous integration and delivery, significantly enhancing software release efficiency.

Continuous integration ensures immediate building and unit testing when developers submit new code. This verifies whether the new and existing code integrates seamlessly based on test outcomes.

Continuous delivery focuses on deploying integrated code into runtime environments, building on continuous integration practices.

With services deployed on Baidu AI Cloud Container Engine, developers can instantly build, test, and package code following submissions on platforms like Git. After integration, the code can be deployed to pre-release and production environments through continuous delivery workflows.