Design And Performance Optimization Of Online Learning System Based On Linux Clusters

With the acceleration of society informatization, led to the the fast pace of knowledge replacement. Therefore, people must continue to absorb the latest information, and learn the latest knowledge to make they not be falling era. Inefficient traditional school education model is unable to meet the demand for people to learn new knowledge In the fast-paced information age. By adding social interaction, online collaborative learning platform allow users to participate in all aspects of self-learning knowledge, the dissemination of knowledge as well as knowledge discovery, greatly improved the efficiency of learning and enthusiasm for learning. Some of the existing collaborative online learning system has shortcomings in terms of high availability and real-time interactive visualization systems.Firstly, in this paper, by incorporating virtualization technology, virtual teaching rooms and virtual laboratory features are proposed, for the lack of real-time interactive visualization functions for Sakai and other online collaborative learning system. This greatly improves the practical interaction between the participants in the learning process. By building on the Sakai platform with virtualization technology, build an online collaborative learning platform prototype system. By analyzing and comparing, result that the system has come online collaborative learning system compared with several advantage.Secondly, for shortages in Sakai system for high availability mechanisms, Linux-based high availability collaborative online learning platform is proposed. Works were done as follows:1) Joining HAProxy Web server cluster to achieve load balancing.2) Ensuring high availability of HAProxy and MySQL Cluster through Keepalived.3) Achieving a highly efficient, highly scalable Web cluster using Session sharing based on Memchached.4) Building up an efficient database cluster mode by master-slaves architecture.5) Achieving a reliability-based copies of redundant data protection mechanisms for the teaching resource storage reliability requirements.Finally, verify the real-time interactive visualization capabilities based on Sakai’s online collaborative learning system improvements, by algorithms, experimental and qualitative analysis. Verify the improvement overall system performance.