基于内容发布订阅系统中高并发事件匹配算法  

High concurrency event matching algorithm for publish subscribe system

作　　者：张志远[1] 钱玭 ZHANG Zhi-yuan;QIAN Pin(College of Computer Science and Technology,Civil Aviation University of China,Tianjin 300300,China)

机构地区：[1]中国民航大学计算机科学与技术学院,天津300300

出　　处：《计算机工程与设计》2022年第12期3414-3420,共7页Computer Engineering and Design

基　　金：空中交通管理系统与技术国家重点实验室开放基金项目(SKLATM201902)。

摘　　要：基于内容的发布订阅系统中事件匹配至关重要,但现有算法仅考虑单一事件匹配,忽略了多个事件之间的内在联系,在高并发情形下匹配效率可能不满足系统要求。为此,考虑事件之间的内在特征,提出一种面向高并发事件的匹配算法HCEM(high concurrency event matching)。在REIN算法索引结构的基础上,动态调整各事件进入搜索空间的顺序,采用负搜索策略进行事件匹配。实验结果表明,在事件高并发情形下与同类算法相比具有更高效的匹配性能。Event matching is very important in content-based publish subscribe system.But existing algorithms only consider single event matching,and ignore the internal relationship between multiple events.And the matching efficiency may not meet the requirements of the system in high concurrency situation.Therefore,considering the inherent characteristics of events,a high concurrency event matching algorithm HCEM(high concurrency event matching)was proposed.Based on the index structure of REIN algorithm,the order of events entering the search space was dynamically adjusted.And the negative search strategy was used for event matching.Experimental results show that the proposed algorithm performs better than other similar algorithms in the case of high concurrency situation.

关 键 词：发布/订阅 事件匹配 谓词索引 多维 高并发 组合优化 实时分发 

分 类 号：TP393[自动化与计算机技术—计算机应用技术]