Internet technology: SAE helps XMiles improve microservices R&D efficiency

Challenge

Benefit

Most casual games have their own lifecycle. During the lifecycle of a game, business traffic during peak and off-peak hours is obvious. For example, the volume of traffic during the day is much higher than the volume of traffic at night. Traffic surges during the day occur at specific points in time and traffic surge at night occurs at 20:00. Almost no traffic exists from 2:00 to 6:00 in the morning, but the service cannot be stopped. The first time a game is released or each time an operational activity occurs, a large number of new customers log on to the game. The backend service must quickly respond to the changes in traffic volume. Therefore, XMiles requires a computing platform that supports auto scaling.

SAE provides multiple types of auto scaling policies to perform scaling operations within seconds. SAE can ensure business service-level agreement (SLA) when burst traffic occurs. SAE provides multiple policies and solutions such as image pre-download, DADI-based image acceleration, cold start acceleration for Java applications, and multi-threading during Java application runtime. This helps you deploy applications, run applications, and perform auto scaling with high efficiency.

To improve service performance and quality, most casual games are stateless and can be split into different service modules, such as chat, red packets, backpacks, upgrades, user data acquisition, video processing, and advertising. In this case, the microservices architecture is suitable for business deployment.

SAE supports mainstream applications such as Spring Cloud, Dubbo, HSF, and Web. SAE is seamlessly integrated with Microservices Engine (MSE) and provides comprehensive service governance capabilities. SAE allows you to use code packages and images to deploy applications without the need to understand the underlying details of Docker images or Kubernetes. You can quickly get started with SAE.

During the release process of a game, multiple new functional modules need to be iteratively added and need to be frequently released and upgraded.

After you create an application, you can upgrade the application by using the canary release method or phased release method. If an issue occurs on the new version, you can roll back the application to an earlier version with a few clicks to ensure the security and stability of the application. SAE is seamlessly integrated with Application Real-Time Monitoring Service (ARMS). You can monitor and diagnose applications in a visualized manner. You can also analyze, troubleshoot, and resolve online issues and configure alerts.

The preceding issues can be fixed by using ECS instances or self-managed Kubernetes clusters. However, these resources require high O&M costs. Manual scaling cannot efficiently handle rapid changes in the volume of traffic. In most cases, a large number of idle resources exist.

You can focus on your core business and the development of business logic without the need to fix the issues that are related to underlying resources, such as cluster availability, capacity, and performance issues. SAE adopts an integrated and fully managed console that provides out-of-the-box lifecycle management and observability features for applications and jobs. This helps you troubleshoot issues with high efficiency. SAE also supports multi-vSwitch deployment to implement cross-zone disaster recovery.