In recent years, increasing effort has been made by the cluster and grid computing community to build object-based Distributed Shared Memory systems (DSM) in a cluster environment. In most ofthese systems, a shared object is simply used as a data-exchanging unit so as to alleviate the false-sharing problem, and the advantages ofsharing objects remain to be fully exploited. Thus, this paper is motivated to investigate the potential advantages ofobject-based DSM. For example, the performance ofa distributed application may be significantly improved by adaptively andjudiciously setting the size of the sharedobjects, i.e., granularity. This paper, in addition to investigating the advantages of sharing objects, particularly focuses on observing how the performance of a distributed application changes with varied granularity, obtaining the optimal granularity through curvefitting, studying the factors that affect the optimal granularity, and predicting this optimal granularity in a changing runtime environment.