Serverless App Engine (SAE) allows you to quickly migrate applications to the cloud. You can configure flexible application start and stop settings to reduce resource and O&M costs. This topic describes how SAE helps Pumpkin Film quickly migrate business data to the cloud.
Background information
Founded in 2015, Pumpkin Film is a vertical video app that delivers high-quality content of movies and TV series. It offers differentiated content operation services to middle-class elites.
Challenges
Pumpkin Film runs based on Alibaba Cloud Elastic Compute Service (ECS). Pumpkin Film also uses other Alibaba Cloud services to build its function modules, such as infrastructure, middleware, database, big data, and security. However, the ECS-based O&M architecture of Pumpkin Film faces increasing challenges as the business grows. The following table describes the challenges faced by Pumpkin Film. The table also describes the benefits of SAE.
Challenge | Benefit |
Challenge | Benefit |
Low scaling efficiency: During peak hours, Pumpkin Film must purchase temporary servers and deploy the servers one by one. The process is time-consuming, and the service level agreement (SLA) cannot be ensured. | The auto scaling feature of SAE integrates with the throttling and degradation features of Application High Availability Service (AHAS) to ensure business continuity in high-traffic scenarios. SAE can scale out applications within approximately 15 seconds. SAE also provides acceleration solutions for the startup of images and Java applications. |
High system maintenance costs: The O&M operations on traditional clusters are complex. The related engineers of Pumpkin Film must be skilled in writing Lua or Ansible scripts, and be familiar with cloud service configurations that are related to networks, monitoring, and O&M. Pumpkin Film did not hire dedicated O&M engineers. As a result, a large amount of developer resources were consumed. The original business system has insufficient monitoring capabilities and requires end-to-end overall monitoring. As a result, the stability of the business system cannot be ensured. | SAE is integrated with Application Real-Time Monitoring Service (ARMS) Basic Edition and automatically installs an agent to provide end-to-end application monitoring and diagnostics capabilities. You do not need to spend a large amount of time in troubleshooting issues. You can use the throttling and degradation features of AHAS on the SAE side to enhance the stability of your microservices system in high-traffic scenarios. |
Difficult capacity planning and low resource utilization: Pumpkin Film is subject to obvious traffic fluctuations, with peak hours concentrated in the evenings, holidays, and after the release of hot movies. Issues such as inaccurate capacity estimation may occur. In most cases, Pumpkin Film retained servers based on traffic for a long period of time, and the resource utilization is relatively low. | SAE provides auto scaling policies to help you handle traffic changes and ensure business stability. You can configure scheduled auto scaling policies and metric-based auto scaling policies. Metrics such as CPU, memory, TCP connections, QPS, and RT are supported. You do not need to retain fixed resources for a long period of time. SAE can scale applications based on the actual business volume to improve resource utilization. You are billed on a minute basis. |
Solution
The following figure shows the solution that is used by Pumpkin Film. 
SAE is a serverless application-oriented platform that provides low barriers to perform cloud migration. SAE hides the underlying details of Kubernetes clusters and provides application-oriented and microservices-oriented UI and API operations that can be directly called by PaaS layer users.
Smooth migration: The new system of Pumpkin Film uses the Spring Boot framework that supports smooth application migration by using WAR packages and JAR packages. The cost of accessing SAE is low.
Reduced O&M workloads and costs: SAE provides the auto scaling feature that helps Pumpkin Film reduce O&M workloads and costs.
Stable business: SAE provides a comprehensive monitoring and alerting system and automated O&M capabilities. The O&M engineers of Pumpkin Film can use SAE to identify and resolve system bottlenecks at the earliest opportunity, and configure auto scaling policies during peak hours to ensure system stability.
Effect
Higher scaling efficiency
SAE automatically scales applications by adjusting the number of instances based on the traffic volume when traffic fluctuations occur. After Pumpkin Film used SAE, Pumpkin Film no longer needs to estimate resource capacity and has reduced hardware costs by approximately 40%.
Faster release speed
SAE supports application deployment by using integration and continuous deployment (CI/CD) tools and the Alibaba Cloud Toolkit plug-in. Pumpkin Film can deploy on-premises applications to the cloud with a few clicks. This facilitates remote debugging.
Effortless O&M
SAE is a maintenance-free service. The O&M process is simple and effortless. For example, if a developer of Pumpkin Film receives an alert notification, the developer needs to only log on to the SAE console. When the developer starts to clear the alert, the O&M process ends. The O&M process provides higher efficiency than manual maintenance.
More efficient troubleshooting
The built-in monitoring capabilities of SAE help Pumpkin Film prevent time-consuming troubleshooting.
