面向服务组合的跨组织工作流性能预测与分析*

摘要

Service‐Oriented Architecture(SOA) promotes the combination of workflow technique and service composi‐tion ,which provides important technical support for Cross‐organizational workflow applications .According to data of Quali‐ty of Service(QoS) detected at runtime ,an analysis approach based on time series is proposed using Autoregressive Integrat‐ed Moving Average(ARIMA) model to predict dynamic performance of workflow systems .When the predicted value beyond or below the setting threshold ,the issues can be analyzed and the service causing performance failure can be found ,to sug‐gest more suitable recovery strategies for service composition .The results of simulation experiment show the effectiveness of the proposed approach .%面向服务的体系结构促进了工作流技术与服务组合技术的结合，为跨组织业务应用开发提供重要技术支撑。针对实时监控到的服务质量属性数据，提出一种基于时间序列的工作流性能分析方法，采用综合自回归移动平均模型来预测工作流的动态性能。当预测到性能值超标时，分析导致工作流性能异常的主要服务，有利于为组合服务提供更加合理的优化服务调整策略。实验结果表明所提出方法的可行性。