Cloud Gaming Server Segmentation for Scalable Multiplayer State
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Conventional client-server networks for multiplayer video games are complex and costly to maintain, making it difficult to scale gameplay to a large number of geographically-distributed players and requiring different development approaches for single-player and multiplayer support.
Innovation Solution
A cloud gaming system utilizing a messaging bus and interconnected servers that form ad-hoc peer-to-peer networks between game simulation instances, allowing players to communicate and share state information within and across regions of a game world without a centralized server, enabling decentralized multiplayer gameplay.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If a centralized server is used to coordinate game state for all players, then game state coordination is achieved, but system complexity and maintenance cost increase significantly
Solution Approach 1:
The game world is divided into multiple regions, and each region has its own dedicated server that manages game state coordination independently. This segmentation allows the system to handle large numbers of players without requiring a single complex centralized server, thereby reducing overall system complexity while maintaining reliable game state coordination within each region.
2Adaptability or versatility
If a centralized server coordinates all players, then multiplayer gameplay is enabled, but scaling to large numbers of geographically-distributed players becomes difficult
Solution Approach 1:
The system divides the player base into multiple regions, each managed by its own server. This allows the system to scale to large numbers of geographically-distributed players by distributing the coordination load across multiple regional servers rather than attempting to coordinate all players through a single centralized server.
Solution Approach 2:
The system introduces a spatial dimension to server architecture by organizing servers geographically according to player locations. This dimensional organization allows efficient scaling to large numbers of players by matching players with regional servers based on their geographical proximity, thereby improving scalability.
3Adaptability or versatility
If different development approaches are implemented for single-player and multiplayer support, then both modes are supported, but development complexity increases
Solution Approach 1:
The game simulation instance is designed to be universal, capable of operating in both single-player and multiplayer modes without requiring substantively different development approaches. The instance can dynamically adapt its behavior based on the presence and number of players, allowing the same codebase to support both game modes effectively.
4Reliability
If player input is communicated through a centralized server, then coordination is achieved, but communication efficiency decreases
Solution Approach 1:
Player input communication is segmented by region, with each regional server handling input coordination only for players within its own region. This reduces communication time by eliminating the need for all player inputs to traverse a centralized server, while maintaining reliable coordination within each regional segment.
Data Source
Figure 1
Figure 2
Figure 3
AI summary
A cloud gaming system (100) includes a messaging bus (108) and a plurality of servers (104) that are interconnected via the messaging bus and communicatively coupled to one or more client devices (106). Each server (104) executes a game simulation instance (208) for each of one or more players associated with one of the client devices (106). Each game simulation instance (208) associated with a player within a specified region (210) of a game world (202) communicates player input by the player to one or more other game simulation instances (210) via an ad-hoc peer-to-peer network (212) formed between multiple game simulation instances (208) associated with players within the specified region (210). Each game simulation instance (208) associated with a player within the specified region (210) also calculates a current state for the player based on the player inputs received via the ad-hoc peer-to-peer network (212).