How Tencent Cloud Powers Global Same-Server Gaming: Architecture & Best Practices

With the rise of mobile internet and global social gaming, worldwide same‑server (global‑server) games like Clash Royale have become extremely popular. Tencent Cloud leverages its global infrastructure and VPC network to offer a complete solution for deploying such games worldwide.

Typical characteristics of a global same‑server architecture include:

Centralized core data deployment: User accounts, core game data, leaderboards, etc., are stored in a central node to ensure consistent interaction across regions.

Regional player proximity: Access, logic, and cache servers are placed near users to improve latency and allow local storage of non‑global data.

Inter‑regional data connectivity: Regional servers communicate with the central server via dedicated lines for cross‑region gameplay, reducing latency and packet loss.

Deployment framework diagram:

1. Service Nodes Worldwide, Centralized Databases

Tencent Cloud operates 18 global service nodes across five continents. Game database servers can be deployed in any of the major data centers based on target player regions (e.g., North America, Southeast Asia).

Selection criteria include product‑side coverage needs and network quality of overseas IDC locations.

Network quality map of over 100 countries to Tencent Cloud IDC (reference only):

North America IDC:

Hong Kong IDC:

For games targeting European players, choose the North America IDC; for East Asian and Southeast Asian users, select the Hong Kong IDC; for primarily Southeast Asian audiences, the Singapore IDC is optimal.

2. Regional Player Access

Due to varying latency from over 100 countries to the central server, players are accessed via multiple regional entry points using Tencent Cloud IDC locations and DNSPod intelligent DNS routing.

Player data can be cached regionally and periodically synchronized back to the central database to avoid real‑time cross‑region reads.

Tencent Cloud provides domestic data centers (Beijing, Shanghai, Guangzhou, Shenzhen) and overseas points (North America, Europe, APAC, South America). Cache services such as Redis or MongoDB can be used to simplify deployment.

Example: Guangzhou regional deployment

Domestic players connect through Guangzhou, deploying access, logic, and cache services locally, then linking to the global servers via dedicated lines.

Note: Not every region requires a full set of game servers; some can use proxy forwarding to improve access speed while controlling costs.

3. High‑Speed Global Network Interconnect

The main challenge for global same‑server games is reducing latency for cross‑region PVP. Network quality varies among thousands of ISPs worldwide, and additional hops increase delay.

Tencent Cloud offers several optimization solutions:

（1）Inter‑Region VPC Connectivity

VPC interconnects create dedicated high‑speed links between regional access servers and the central server, dramatically lowering latency compared to public internet. Setup requires two steps in the console:

Create a peering connection.

Configure routing tables on both ends.

After creation, the connection status can be viewed in the console.

For example, a Hong Kong‑to‑North America VPC link can keep latency around 200 ms, satisfying requirements for games like COK or COC.

（2）Overseas POP Acceleration

Beyond IDC interconnects, Tencent Cloud provides POP points in Europe, South America, Africa, etc., to reduce cross‑ISP routing and bypass export restrictions.

When the game is deployed in a North America IDC, European users can connect via the nearest POP point, then traffic is tunneled back to the North America IDC.

Latency from Europe to North America after POP acceleration is typically under 100 ms.

Summary

Centralized global data: Store accounts, leaderboards, etc., in a central IDC (e.g., North America) to cover target regions.

Regional access points: Deploy access services in major regions (Guangzhou, Singapore, North America) and use DNSPod intelligent routing and load balancers for proximity.

Minimize cross‑region data exchange: Keep players within the same region when possible; use regional logic servers and cache databases (Redis, CDB) to lower latency and operational cost.

Network optimization: Employ VPC dedicated lines or POP acceleration to achieve high‑speed interconnects and ensure smooth global same‑server gameplay.