Online Reliability Time Series Prediction Via CL-ROP For Service Systems

With the development of Web service technology,more and more enterprises choose to publish their own services on the Internet.However,with the increasing demands of users,it is difficult for a single service to meet the users' requirements.Service system integrates multiple services based on SOA architecture so that it can satisfy the uses' complex demands.The component services of service system is loosely coupled.However,due to the dynamic running environment,the performance of each component service(including reliability)will always fluctuate,may even bring cascading effects,and then cause the entire Service system to shut down.Online reliability prediction for service systems that ensures the runtime quality poses a major challenge and attracts growing attention.In order to build a highly reliable service system and ensure the stable operation of the service system,reliability prediction has become popular in recent years.To solve this problem,the main challenges are as follows:component services run in a dynamic environment,and the parameters used to carry out reliabili-ty prediction are difficult to obtain.This paper analyzes the historical reliability time series of component services,and predicts the reliability in the near future.To guarantee the stable and continuous operation of systems,we propose a online reliability time series prediction method basing on Convolutional Neural Network(CNN)and Long Short Term Memory(LSTM)which we call it CL-ROP to predict the reliability of service systems in the near future.We conduct a series of experiments on a dataset composed of real Web services and compare with other competitive approaches.It is proved that our CL-ROP method has advantages with other approaches.