Using Coherence To Extend The Session Replication Function Of WebLogic Clustering And Implementation

Cloud computing is important innovation in the field of contemporary information, it appears at this stage has its inevitability, is to address the information society "three big" challenge (the "big user", "big data" and "big system"). Six technical characteristics of Cloud Computing can be summarized as:elastic, transparent, modular, versatile, multiple rentals, and dynamic. It can be seen that the six characteristics of current business IT systems are widely used in J2EE clustering technology has a universal common.Clustering refers to more than one computer in the traditional sense, redundant connectivity for users to form a single high-availability system. Cluster computing is to become network two or more computers to take full advantage of parallel processing performance technology. With the arrival of the cloud computing era, simple clustering technology has been unable to meet the growing "user" and "system" process requests. Future a major development directions of cluster computing is grid computing. Structurally, grid computing is a distributed computing paradigm, by dispersing in the compute nodes connected together in a network to form a virtual computer with a super performance, providing end users with a more powerful computing and storage capacity.As one of the important technologies supporting cloud computing, J2EE middleware clustering technology will become one of the core components of the enterprise cloud platforms now and in the future. But with the proven J2EE middleware products and after careful consideration of the cluster framework will bring tremendous economic value for the enterprise and higher operational efficiency.This article will focus on how to present the mainstream J2EE middleware product-WebLogic, and gradually became popular grid computing product-Coherence, as the basis for an e-commerce company internal collaboration system design of a load-balanced, highly available, and has powerful scalability of J2EE clustering architecture. In order to make the system has good scalability and ease of management, the system architecture will adopt a separate Array of hardware load balancing devices, Web servers, WebLogic Web Layer cluster, WebLogic EJB layer cluster and Coherence Cached cluster, allows users to flexible configuration based on system characteristics of each cluster node number. Finally, we will compare the system client user requesting Http session information is cached in a WebLogic cluster and Coherence cluster in higher concurrent user access impact on system performance.Test results show that the system, the system the client user after the request Http session information is cached on the Coherence of clusters, improvement of system performance will be more obvious, especially in the case of the large number of concurrent users accessing a whole improves the performance of J2EE clustering.