基于硬件加速的高速数据流连续实时聚集查询

近年来,动态数据流环境下的聚集查询正成为一个热点研究问题.目前的相关算法主要是采用近似技术,以牺牲精度来换取处理速度的提高.然而,在高速数据流环境下,处理速度仍然难以满足需求.软硬件协同的高速数据流处理技术逐渐引起人们的关注.提出了一种基于硬件加速的高速数据流聚集查询方法,充分发挥硬件在处理速度上的优势和软件在灵活性方面的长处.算法是增量的,也实现了多窗口资源共享.最后,给出了算法的复杂度分析并实验验证了方法的有效性.

Continual Aggregation Queries over High Rate Data Streams Based on Hardware Acceleration

Recently there has been a growing interest in aggregation queries for scenarios in which data streams arrive at very high rates and a data stream system is registered with many simultaneous queries. In order to dealing with the huge amounts of data and increasingly stringent response-time requirements, Most existing work in this area has to adopt approximate technology which sacrifice aggregate veracity. But in the environment of high rate data streams, the processing rate still cannot satisfy requirements. So query processing based on hardware-software codesign has recently emerged as a viable solution for dealing with high rate data streams.In this paper, We propose a kind of novel aggregate query algorithms based on hardware-software codesign, which incorpo- rate hardware advantage in processing rate and software long suit in agility. Many incremental computation approaches and resourue sharing techniques in sliding-window aggregations are introduced.Lastly,time cost of the algorithm is analyzed and the experiment show the feasibility and effectiveness of the approach.