Storm Based Research On Optimization Of Middle Sized Web Learning System

With the rapid development of information technology,big data age is coming.Big data revolution is sweeping over almost every field,education has no exception too.With the development of educational informatization,some new digital education models have been proposed,many e-learning systems have been built.In certain middle sized Web learning system,many issues about high delay of responses,missing of system data arisen because of the great increase of user volume.These issues will greatly impact the user's experience.It's a big challenge to solve these issues without modifying original system framework.After researching about Storm distributed computing framework found that Storm is not only able to process data in high-efficiency but also insure the data security.And Storm has been applied in many fields already like,financial system,power system,IOT system et.al.In this paper,we will try to solve the issues in middle sized Web learning system by introducing Storm distributed computing framework to improve original system.The work of this paper includes mainly four aspects as followed:First,undertaking a research on big data theories and big data processing technologies,we analyzed the advantages in many big data processing technologies like,Hadoop and Storm.Second,we analyzed the issues in middle sized Web learning system and its causes.Then this paper proposed a Storm-based improvement project for middle sized Web learning system by using Flume technique for data collecting,Kafka and Memcached technique for the data buffering and Storm technique for real-time data computing and analysis.Third,the Storm-based improvement project for middle sized Web learning system has been built on a real Web learning system.Fourth,the testing about the proposed Storm-based improvement project has been made.By comparing the proposed improvement project and the original Web learning system,we found that the proposed improvement project can solve the issues about high delay of responses,missing of system data effectively.And the project is able to provide high-quality data services.The major novelty of this paper is proposed a Storm-based improvement project for middle sized Web learning system to overcome the issues about high delay of responses,missing of system data in original system.And the improvement project is independent of original system framework.