Performance analysis of Optimistic Concurrency Control schemes with different rerun policies

Abstract

Concurrency Control (CC) is critical to the performance of transaction processing systems. Optimistic CC (OCC) offers an alternative to locking but is vulnerable to resource contention due to transaction aborts. With sufficient buffer, data blocks referenced by aborted transactions can be kept in memory and be available for access during rerun, thus greatly reducing the abort probability during rerun. Consequently, the pure OCC scheme, which only aborts a transaction at its commit time, can do better than aborting transactions as soon as conflict is detected (broadcast OCC). To further exploit this phenomenon, hybrid CC schemes are devised which employ a different CC scheme to handle rerun transactions. These include switching to static or dynamic locking (static and dynamic hybrid OCC schemes, respectively) or to broadcast OCC during rerun, while doing pure OCC for the first run. In the high data contention environment, where locking is inferior to OCC, we find that the static and dynamic hybrid OCC can do better than OCC. The elimination of IO during rerun reduces the data contention faced by the rerun transactions and makes locking preferable to OCC for rerun transactions. Substantial performance improvement can thus be achieved.