How Cloud‑Native Architecture Powers Global Game Publishing at Lingxi Interactive

Lingxi Interactive is a game development and operation company that focuses on global publishing, offering strategy (SLG), card, MMORPG, and open‑world sandbox titles. By the end of 2020 its flagship Three Kingdoms Strategy was launched overseas on a Kubernetes (K8s) platform.

Globalization Challenges

Fragmented overseas markets require flexible release strategies, multi‑region synchronization, and a shared service ecosystem for operations, publishing, and payment.

High‑latency, device diversity, and long‑haul network issues make root‑cause analysis difficult, compounded by language, timezone, and data‑management barriers.

Frequent attacks on overseas nodes demand extensive security resources, while traffic spikes and lack of unified metering hinder cost control.

Technical Strategy: A Unified Global Infrastructure

The team decided to build a generic, flexible global infrastructure platform that can be constructed once and reused everywhere, improving resource efficiency and reducing development and operations costs.

Key constraints in the gaming industry made this difficult:

Strong isolation : Different games and server partitions lead to fragmented resources that cannot be unified by simple standardization.

High performance & customization : Games need long‑lived connections, stateful services, and low‑latency responses, conflicting with stateless cloud‑native designs.

Technical debt : Limited manpower and tight development cycles prevent teams from tackling complex infrastructure projects.

Three‑Layer Cloud‑Native Stack

Kernel layer : Uses Alibaba Cloud Container Service for Kubernetes (ACK) as the base, defining business models and application architecture through standardized container scheduling.

Core layer : Provides compute, storage, and network resources, ensuring flexibility and scalability.

Extension layer : Adds plug‑in or custom SaaS services to adapt to various business service types.

Automation & Game‑Specific Adaptations

Previously, capacity expansion required manual steps involving monitoring, ticketing, server procurement, tagging, and publishing, often taking an hour and risking traffic snowballing. By leveraging ACK’s high‑performance, scalable container management, Lingxi automated the entire pipeline (IaC + GitOps): monitoring triggers a scaling policy, resources are provisioned, initialized, labeled, and rolled out via CI/CD with a single click.

The team also integrated the open‑source OpenKruiseGame project to enable in‑place hot updates, shortening release cycles and improving operational efficiency.

Observability and Monitoring

To guarantee low‑latency, high‑availability services, Lingxi combined Alibaba Cloud Log Service (SLS), CloudMonitor (CMS), network quality analysis, and ARMS for a full‑stack, end‑to‑end observability platform. This monitors infrastructure and application metrics, tracks performance bottlenecks (e.g., payment latency, matchmaking response), and closes the feedback loop for continuous improvement.

FinOps and Cost Optimization

Facing multi‑region deployments and traffic spikes, the team applied FinOps principles to visualize costs, tag resources automatically, and trace expenses throughout the lifecycle. ACK’s elastic scaling aligns resource usage with active player counts, avoiding waste during low‑traffic periods. Metrics such as per‑player operational cost guide cross‑team cost‑saving initiatives, achieving over 40% cost reduction in two years.

Results and Outlook

Infrastructure failure rate dropped from double‑digit to single‑digit percentages.

Key operational tasks (especially scaling and maintenance) became >97% faster.

Overall cloud cost decreased by more than 40% while resource utilization rose sharply.

User experience improved through rapid issue detection and resolution.

Going forward, Lingxi plans to further refine its cloud‑native base, address international resource supply and security challenges, and continue leveraging ACK’s FinOps capabilities for finer‑grained cost governance.