论文部分内容阅读
提出了基于直方图的降载策略,能很好地减缓在过载发生时系统性能的下降.降载的目标在于删除过载数据的同时尽可能地保持数据流的特征.为了处理大量被延迟的数据,构建了一种塔形矩阵的数据存储结构,利用其对过载数据分桶,每桶提取一个代表数据并删除该桶中其余数据,将每个桶的代表数据组成新的数据流参与查询操作.实验结果表明:这种降载方法能有效减少系统负担,生成的新数据流参与数据流查询后所得查询结果错误率较低,其性能优于其他已有算法.
A histogram-based load shedding strategy is proposed, which can effectively reduce the system performance degradation in the event of overload.The goal of load shedding is to delete the overloaded data and keep the characteristics of the data stream as much as possible.In order to deal with a large amount of delayed data , A tower matrix data storage structure is constructed. By using it to parse the overload data, each bucket can extract one representative data and delete the remaining data in the bucket. The representative data of each bucket forms a new data stream to participate in the query operation The experimental results show that this method can effectively reduce the system load, and the new data stream generated by the data stream query results in a lower error rate and better performance than other existing algorithms.