Analysis of application partitioning for massively multiplayer mobile gaming

  • Authors:
  • Madan Kumar M. M;Amit Thawani;Sridhar V;Y. N. Srikant

  • Affiliations:
  • Satyam Computer Service Ltd, Bangalore, India;Satyam Computer Service Ltd, Bangalore, India;Satyam Computer Service Ltd, Bangalore, India;IISc Campus, Bangalore, India

  • Venue:
  • Proceedings of the 1st international conference on MOBILe Wireless MiddleWARE, Operating Systems, and Applications
  • Year:
  • 2008

Quantified Score

Hi-index 0.01

Visualization

Abstract

Mobile devices offer the opportunity to play games anywhere anytime. Moreover, networked games allow individual players to interact with other people and to participate in a larger gaming world. Improved network and upgraded software's available on mobiles have given enough scope for massively multiplayer mobile games. An inherent problem is efficient utilization of resources when a numbers of people are playing games in real time. Gaming infrastructure mostly involves gaming servers and it is likely for a gaming server to run short of resources under peak load conditions resulting in degradation of game play. Under this situation possible solutions would be to replicate server to handle more load, increasing the bandwidth, or to maintain different connections with other servers. Since load on server is not likely to happen often, replicating of server infrastructure prove to be costly. To handle such situation possible solution is to partition the game application and off load some of the processing onto either client/server depending on the availability of resources provided the other has sufficient processing bandwidth available. In this paper we address the problem of providing a good gaming experience on mobile devices when server is short of resources. Our approach considers the game which follows client server model and is based on partitioning the game application. We model the game and represent the game as a graph and partitioning game application problem reduces to graph partitioning approach.