| The O2O business model combines offline services with online internet platforms,providing new ideas and opportunities for the development of traditional service industries in capital flows,information display and promotion.The reservation O2O scenario in this paper has a wide range of applications,such as Helijia and Meidaojia in China,and Care.com domestic service,Soothe massage services abroad.The module of O2O system is relatively complex,and the method of de novo development is inefficient,redundant and difficult to maintain.There are many modules with similar functions in different O2O systems,so a universal O2O platform can effectively reduce the development and maintenance workload of O2O systems.At the same time,due to the limited number and time of service personnel,online service of reservation-based considers more of time and space,and each kind of specific scenario has unique demand.Taking into account the characteristics of this scenario,this paper is devoted to building a reservation-based O2O service system cloud platform that allows multiple services to intervene and supports flexible configuration.The system allows custom configuration of a series of configurations and parameters to accomplish a variety of individual needs.This paper first analyzes the system roles and business requirements in the reservation-based O2O scenario and finds the similarities and differences.The system store specific configuration information and change history via time-linked list sequences to meet multiple-shop and configurable needs.The system is based on micro-service architecture development,each module such as configuration,order work together through remote process calls and message system to complete the function.The system develops in a layered approach and is divided into six layers:access layer,load balancing layer and data layer,etc.At the same time,the system uses a variety of ways to optimize and improve in order to cope with the high concurrency requirements generated by multiple shop access.For example,the maintenance of order status is implemented by the chain of responsibility model and the finite state automaton.The GeoHash algorithm is used for space management,and the idea of "time slice" is used for time management.Finally,the system is tested from two aspects of function and performance.As a unified cloud platform,the system is suitable for access of various reservation-based O2O scenarios,develop once and benefit for a long time.As of the end of November 2017,the system has access to two shops,all with good business results.Thus,the system does have a certain effect on improving the efficiency of development and maintenance and promoting the development of personalized O2O. |
PDF Full Text Request