Long Distance Redundancy Reduction in Thin Client Computing

Abstract

An efficient way to improve the performance of a thin client system is to reduce the redundant screen update data sent across networks. While various optimization efforts have been made in this way, these optimization techniques cannot efficiently reduce long distance redundancies that may contain lots of network traffic. Long distance redundancies can be potentially reduced by a LZ compression algorithm with a large history buffer. However, a flat extension of the history buffer is not flexible and has scalability issue. In this paper, we present a different way to extend the history buffer and a history extension scheme based on it, aiming at minimizing long distance redundancies. We empirically studied the effectiveness of our scheme on some screen updates generated by one of the most bandwidth-efficient thin client system, Microsoft Terminal Service. The results show that this scheme can reduce up to 9.39-23.3% network traffic for the tested data with a history buffer of 512K bytes. This scheme costs only a little additional computation and memory resources.