How UCloud Powered Rapid Mask-Reservation Systems During COVID-19

UCloud Public Cloud Rapid Response

The pandemic made mask supply a critical issue, prompting local governments to launch online reservation systems to reduce crowding and cross-infection. UCloud partnered with the data bureaus of Yuyao, Beilun, and Haishu districts in Ningbo, deploying mask reservation platforms within hours and iterating them over a few days.

Approach: Leveraging Mature Public‑Cloud Components

To meet tight deadlines, the team modularized the system into front‑end, back‑end, and database components, applying DevOps principles and CI/CD pipelines for continuous integration and delivery. Mature UCloud services—UK8S for Kubernetes, UDB for MySQL, UES for log collection, and UHost with GitLab for code management—were used to accelerate development and deployment.

The deployment architecture included three UK8S namespaces (pre‑production, production, performance‑test) with corresponding LoadBalancer or ClusterIP services, ELK for log aggregation, and separate filebeat/sidecar agents for application logs.

DevOps: Full‑Stack Capabilities

With a small volunteer team, the developers performed all coding on the master node, tagging releases for integration and deployment. Database instances were split into development and production configurations, and the application was deployed as a Kubernetes Deployment across the three namespaces.

CI/CD: Boosting Development and Operations Efficiency

A complete CI/CD pipeline was built using GitLab and GitLab Runner. The workflow proceeded from unit testing to commit, push, tag, and manual deployment to production after pre‑release verification. This enabled rapid iteration—over 50 versions across three environments in three days—supporting frequent business rule changes such as address‑based or ID‑based mask distribution.

Public‑Cloud Operations Experience Ensuring System Stability

UCloud provided 24/7 monitoring, capacity planning, and elastic scaling. During peak periods, traffic exceeded 200 Mbps in Beilun and 800 Mbps in Yuyao, comparable to small internet companies. The cloud’s massive bandwidth and auto‑scaling capabilities handled the load, allowing 10,000 masks to be reserved within 30 seconds.

Conclusion

The rapid, cloud‑native deployment demonstrated that a small, well‑coordinated team can leverage public‑cloud services to deliver high‑availability, high‑throughput applications during emergencies. UCloud’s flexible infrastructure, extensive bandwidth, and robust monitoring proved essential for supporting pandemic‑related online services.