High‐Performance Computing on Clusters: The Distributed JVM Approach
Wenzhang Zhu, Cho‐Li Wang, Weijian Fang, Francis C. M. Lau · 2005
this memory model. Before a thread releases a lock, it must copy all assigned values in its private working memory back to the heap which is shared by all threads. Before a thread acquires a lock, it must flush (invalidate) all variables in its working memory; and later uses will load the values from the heap