Although the reliability of the composition of web services has attracted much research works about it, but an important facet of it – MTTF (Meantime to Failure) has not been given enough considerations. The research presented in this paper intends to fill this gap by illustrating on an example of composite web service how the redundant system works and what kinds of practical results can be derived. The main contributions of this work includes: First, provide the concept of MTTF of composite web service, which is little concerned in previous works. Second, describing the calculation method of MTTF of composite web based on the workflow composition pattern. Third, presents the quantitative analysis of MTTF of composite web service for non-redundant services, part-redundant services, and all-redundant services based system. And we show that by an experiment, to achieve the higher reliability of a system, it is necessary to decrease the failure rate and increase the repair rate in addition to providing redundant system.