分布式数据流系统通信有效性研究综述

A Survey of Communicate Efficient Algorithms of Distributed Data Stream

在分布式数据流环境中,系统的通信带宽是一种瓶颈资源。在保证查询精度的前提下,有效地减少网络中数据流的传输量是解决这一问题的重要途径。通过分析现有的分布式数据流处理算法,总结出一个通用处理框架,以减少数据流的传输量。通用处理框架包括三个方面:最小化信息传输、使用数据流摘要表示完整信息以及通过预测维持系统的稳定性。

In distributed data stream systems, the available communication bandwidth is a bottleneck. Thus, effective solutions have to be communication efficient, and provide continuous, guaranteed quality queries. Aimed at this practical problem, based on the analysis of current distributed data stream processing algorithms, a general tracking frame has been summarized, which based on three principles of minimizing global information, using summaries to capture whole data streams, and using prediction model to seeki...