Research on sliding window join semantics and join algorithm in heterogeneous data streams

Sliding windows of data stream have rich semantics, which results all kinds of window semantics of different data stream, so join semantics between the different types of windows becomes very complicated. The basic join semantic of data streams, the join semantic of tuple-based sliding window and the join semantic of time-based sliding window have partly solved the semantics of stream joins, but the heterogeneity of sliding windows is difficult to be solved. In this paper we present the join semantic model based on matching window identifies for joining of multi-data stream. We make use of window identifies to shield the difference of window attribute, window size, and window slide. In this paper, a sliding window is divided into a number of sub-windows when the newest sub-window fills up it and it is appended to the sliding window while the oldest sub-window in the sliding window is removed. We use the equivalence relation of overlapping sub-window belonging to the adjacent sliding window to reduce the number of join computing. We propose the corresponding algorithm of window join to maintain the window. The theoretical and experimental analysis show that the joining model of window identifies can synchronize multiple data stream. © Wei and Xianxia.