適用於多人線上遊戲動態負載管理之混合同儕式雲端架構

Abstract

近年來由於大型多人線上遊戲(MMOGs)的蓬勃發展，引起了學界和業界的高度興趣。多人線上遊戲需要高度的彈性來面對環境的變化，因此我們利用雲端的特性來與MMOG做結合。在雲端遊戲環境中，我們利用虛擬機器來取代傳統實體遊戲伺服器。由於現今MMOGs主要系統架構為多伺服器架構，虛擬遊戲世界會被切割成數個遊戲區域，每個遊戲區域由一個或多個實體遊戲伺服器負責和客戶端玩家進行遊戲資訊的傳輸及執行。然而，在多伺服器架構下，會因為大多數客戶端玩家感興趣的地圖區域相同，而造成遊戲伺服器負載不平衡。在本論文中，我們在MMOGs中使用同儕式雲端計算架構，使MMOGs在雲端計算的環境下獲得較彈性的資源利用。同儕式雲端計算是一個結合高計算能力、高擴充性、高可信賴，以及分享伺服器資源及資料的新概念。本論文提出一個混合同儕式雲端架構，此一架構可以更適用於大型多人線上遊戲，它改進了現行多人線上遊戲多伺服器架構的缺陷。除了提出一個適用於多人線上遊戲之混合同儕式雲端架構外，我們也針對每一個遊戲伺服器提出多門檻負載管理機制和遊戲伺服器之間的負載管理機制。本篇論文的架構與多伺服器架構相比較，在小於300毫秒的回應時間內，我們的架構比多伺服器架構可多服務10.31%的玩家量。此外，在中高負載時，我們的架構比多伺服器架構少27.9%的錯過期限比率。

In recent years, massively multiplayer online games (MMOGs) become more and more popular. Many researchers, both in academia and industry, are very interested in MMOGs. MMOG environments require a high degree of flexibility to respond to environmental changes, including load change. We combine cloud computing with MMOGs to increase flexibility of resource allocation. In an MMOG cloud environment, we use virtual machines (VMs), instead of traditional physical game servers. A game world is divided into several game regions. Each game region is serviced by at least one VM. However, in the multi-server architecture, loads of regional servers may be unbalanced because there may be some regions that attract more players. Peer-to-peer (P2P) cloud computing is a new approach that combines high computation power, scalability, reliability and efficient information sharing of servers. This paper proposes a hybrid P2P cloud architecture for MMOGs which includes two-level load management, multi-threshold load management for each game server and load management among game servers. It is suitable for players to interact with P2P cloud servers and it avoids bottlenecks of the current multi-server MMOG architecture. Simulation results show that the proposed architecture can support 10.31% more players under no deadline (300 ms) miss compared to the multi-server architecture. The proposed hybrid P2P cloud architecture can reduce the average response time by 20.6% composed to the multi-server architecture under medium to high load through flexible allocation of resources (virtual machines). The proposed architecture also has a 27.9% smaller deadline miss ratio than the multi-server architecture under medium to high load.