Consistency Models for Mobile Games with Cloud Databases

Abstract: Developing games on smartphones with online features requires careful designing in order to scale to a large amount of users. In this thesis, selected databases are evaluated for use in mobile games, to store game data locally, in terms of performance and usability. The databases are compared to the current storage implementation of a game developed for iOS and Android devices. It is found that the databases perform better when reading or writing small parts of a larger data set, and scale well as the data set increases in size. However, the size of the data in the game studied is not large enough for the increase in complexity to outweigh the gained performance. In addition to this, a library is written to simplify synchronizing state between server and client on mobile games.