QoS Aware Real-Time Platform for Massively Multiplayer Online Gaming (MMOG) over RTPS
No Thumbnail Available
Date
Authors
Journal Title
Journal ISSN
Volume Title
Publisher
Saudi Digital Library
Abstract
Recently, Massively Multiplayer Online Game (MMOG) has become very popular; there are thousands of games built to be played online, where hundreds or thousands of users from all around the world can play the game at the same time. The hottest issue in the MMOG is the real-time (RT) service, where the players can share their game status in a real-time manner. The main need for the RT platform is to guarantee the quality of services, the platform should deliver a robust and excellent quality, besides, the service should be loosely couple oriented to support the growth of MMOG applications.
We present a new platform solution to deal with the real-time online gaming using a DDS middleware. An RTinDDS is a real-time platform built based on the DDS middleware. This middleware can guarantee to provide the developers, designers, and end users with a suitable platform to build their online real-time gaming and to deliver it in a robust manner to the end users. The platform also allows designers to make the best decision for a specific situation. The results of the experimental work show that RTinDDS improved the reliability, scalability, throughput, and latency. Chapter 6.
We improved RTinDDS by implementing a new smart algorithm based on the artificial neural network AIRTinDDS. This algorithm used to detect and resolve any issue will occur during the game running to guarantee to deliver a robust platform, the system able to adapt itself to any unexpected behavior. This will effect on many factors such as, sharing the resources fairly among the players, describing the issue to the players and inform them what to do and solve the issue without any break in the game.
We have applied our algorithms RTinDDS/AIRTinDDS to build a game (PMS) over DDS middleware to measure and test the power of RTinDDS/AIRTinDDS, we tested many scenarios and profile to make sure that we covered most cases and circumstances. The results show very good throughput with a very reasonable latency, with the ability to scale the number of players.
