| With the development of information technology, more and more datas burst into our eyes. Due to the defects of current common database on processing these massive data streams with high-speed, more and more people began to focus on how to develop a new processing method to deal with these high-speed, real-time and continuous data stream or event stream.Nowadays, complex event processing (CEP) is one of the event streams processing methods, which uses certain algorithm to filter out valid events, and then detects the complex event and promote the upper layer application to make a corresponding processing. Firstly, this paper made a simple comparison between complex event detection and database system, reflecting the former's advantages on massive event stream processing. Then, we made a brief overview and comparison of complex event detection technology between home and abroad.Based on current CEP technology, this paper presented a directed graph based multiple complex event detection method. It used SASE as events and complex event pattern description language, which is a hot language in CEP. This paper improved the SASE language to process multiple complex events better, designed the improved SASE's compiler system and algorithm to link multiple complex event into a directed graph and put forward shared factor of the directed graph, which could be used as shared performance of these multiple complex events. In this paper, it designed CEP algorithm based on the directed graph, which uses single effect event to traverse the directed graph. When the graph traverse to some special vertex, certain complex event will be detected and ask upper layer application to make corresponding processing.Finally, the simulation experiment made a time complexity comparison of the CEP of different shared factor value. Meanwhile, it compared with the traditional technique based on NFA. By contrast, our method had some advantage in detecting short-lived complex events. |
PDF Full Text Request