基于收益机制发布／订阅系统时间约束保障技术  

An Earning-Based Time-Constraint Assurance Technique for Distributed Publish/Subscribe System

作　　者：马建刚[1] 黄涛[1] 徐罡[1] 汪锦岭[1] 叶丹[1] 

机构地区：[1]中国科学院软件研究所软件工程技术研究开发中心

出　　处：《软件学报》2008年第7期1590-1602,共13页Journal of Software

基　　金：the National Natural Science Foundation of China under Grant No.60573126(国家自然科学基金);the National High-Tech Research and Development Plan of China under Grant Nos.2007AA04Z148, 2007AAO1Z149(国家高技术研究发展计划(863));the National Basic Research Program of China under Grant No.2002CB312005(国家重点基础研究发展计划(973))

摘　　要：发布/订阅系统技术具有异步、松散耦合和多对多通信的特点,有着广阔的应用前景.但是,已有的发布/订阅系统技术不能满足动态环境下有延迟需求的应用要求.针对时间约束问题,扩展了发布/订阅系统的语法,建立了延迟模型,提出了一种基于收益机制的分布式发布/订阅系统时间约束保障技术和使系统获益最大化的调度算法MTEP(maximum total earning priority),其特点是能够满足订阅者和发布者指定延迟约束的需求,通过与订阅者商定的价格和违约成本信息来有效地利用网络带宽,适应网络环境的动态变化.实验结果表明,该调度策略和FCFS(first come first service)、最短时间优先和固定优先级等传统策略相比,可使订阅者接收到的有效事件明显增多,并使系统收益显著改善.The publish/subscribe （pub/sub） system has a wide potential application due to its asynchronous, many-to-many and loosely-coupled communication properties. However the existing pub/sub systems can＇t satisfy the delay requirement of application in dynamic distributed computing environment. In order to solve this problem, this paper extends the syntax of pub/sub system, sets up the delay model, proposes an earning-based time-constraint assurance technique for distributed pub/sub system, and puts forward a scheduling algorithm named as MTEP （maximum total earning priority）. The algorithm satisfies the specified delay requirement for both publisher and subscriber, makes use of available bandwidth efficiently with the price and penalty information agreed with subscribers, and adapts to the dynamic network environment. Experimental results show that the strategy can make subscribers receive many more valid events and improve system earning significantly comparing with the classical FCFS, fixed priority and least remain time priority algorithms.

关 键 词：发布/订阅 事件模型 延迟模型 调度算法 优先级 价格 违约成本 

分 类 号：TP311[自动化与计算机技术—计算机软件与理论]