READ-COPY UPDATE: USING EXECUTION HISTORY TO SOLVE CONCURRENCY PROBLEMS 论文

摘要

Synchronization overhead, contention, and deadlock can pose severe challenges to designers and implementers of parallel programs. Therefore, many researchers have proposed update disciplines that solve these problems in restricted but commonly occurring situations. However, these proposals rely either on garbage collectors [7, 8], termination of all processes currently using the data structure [10], or expensive explicit tracking of all processes accessing the data structure [5, 15]. These mechanisms are inappropriate in many cases, such as within many operating-system kernels and server applications. This paper proposes a novel and extremely efficient mechanism, called read-copy update, and compares its performance to that of conventional locking primitives.