Progressive Hash-Merge Join Algorithm

In data streams or Web scenario at highly variable and unpredictable rates, most fast join algorithms to date rely upon shifting to external join stage as soon as possible when blocked in order to enhance efficiency. But they have trouble with the following issues: the limit of external join and practical executing time. Classical progressive two-way joining technique based on hash, however, fail to deliver acceptable performance in such a scenario where relatively short intermittent delay exists in the gross. We propose a new progressive join algorithm based on hash-merge for improving the query response time, separating one merging transaction into multi-subtask, whose transacting sizes rest with the interval time. Additionally, a refined replacement selection tree and a fine granularity timestamp are applied, which help to make use of finite memory and ensure correctness respectively. Theory and experimental results show that our technique delivers results significantly fast under both reliable and unreliable network.