.png){kind=icon}
.png){kind=icon}
Return
MAP-MIND: An Offline Algorithm for Optimizing Game Engine Module Placement in Cloud Gaming
.png){kind=icon}
OA
AI
DOI：10.1109/ACCESS.2024.3380900.png)
.png){kind=icon}
Original request for help.png){kind=icon}
Share.png){kind=icon}
Save
Abstract
En 中文
Online gaming has seen a significant surge in popularity, becoming a dominant form of entertainment worldwide. This growth has necessitated the evolution of game servers from centralized to distributed models, leading to the emergence of distributed game engines. These engines allow for the distribution of game engine modules (GEMs) across multiple servers, improving scalability and performance. However, this distribution presents a new challenge: the game engine module placement problem. This problem involves strategically placing GEMs to maximize the number of accepted placement requests while minimizing the delay experienced by players, a critical factor in enhancing the gaming experience. The problem can be formulated as an Integer Linear Programming (ILP) model, which provides an optimal solution but suffers from high computational complexity, making it impractical for real-world applications. To address this challenge, this paper introduces two novel heuristic algorithms, MAP-MIND and MAP-MIND*. The MAP-MIND algorithm demonstrates superior performance, achieving near-optimal delay and more than 92% GEM request acceptance in the worst heterogeneous scenarios. The MAP-MIND* algorithm, while slightly under-performing MAP-MIND in terms of delay, proves to be significantly faster, making it a viable alternative for real-world applications with equal GEM request acceptance. The trade-off between the two algorithms offers a flexible approach to GEM placement, balancing performance and computational efficiency.
Keywords:
Cloud gaming
distributed game engines
placement algorithm
Journal
IF:
3.6
Papers:
9.8W
Citations:
29.4W
View more organizations