Abstract
Consistency maintenance in distributed interactive applications (DIAs) is subjected to network characteristics such as limited bandwidth and latency. Predictive contract mechanisms are techniques that compensate for the effect of network latency by extrapolating future entity states from historical records. These approaches trade inconsistency within human perceptual limits for reduced network traffic and latency. This paper explores the use of an information metric to analyse the effect of network latency on remote consistency and thus establishes a novel framework to model predictive contract mechanisms as a lossy information sharing process. Such a perspective facilitates a novel explicit analysis of the trade-off between network traffic and inconsistency.

This publication has 12 references indexed in Scilit: