Study And Application Of Mobile Terminal Cloud Data Access Mode Based On Hadoop

With the rapid development of modern technology, people can query travelknowledge and information through Internet. It impels tourists to prefer to arrangetheir travel according to their personal willingness. So, self-help travel proportion isincreasing year by year. And the tourism information service market becomesbooming nowadays. In addition, the sudden and uncertainty of travel process decidethat Mobile Internet is the main way that tourists query travel information. Mobilecloud computing introduces the concept of cloud computing into the Mobile Internet.On the one hand, it makes full use of the portability of mobile terminals. On the otherhand, it makes up shortages of the mobile terminal in computing and storage capacity.In order to apply mobile cloud computing in the tourism information retrieval service,it’s necessary to propose the mobile terminal cloud resources access patterncombining with the characteristics of mobile terminal and travel behavior.In order to design the mobile terminal cloud resources access pattern based onHadoop, first of all, it’s necessary to deeply understand the basic component ofHadoop, and emphatically analyze the working principle of the two core components:HDFS and MapReduce. The consistency of Hadoop and cloud computing in viewsand key technologies makes Hadoop become a distributed cloud computing platform.Secondly, the cloud resources access pattern is divided into three sections: datastorage strategy, request scheduling algorithm and data response method, and isanalyzed under Hadoop common pattern. Then, combining with the mobile terminalcharacteristics of small computational amount, high concurrency and real-timerequirement, the mobile terminal cloud resources access pattern is purposed based onHadoop. Data storage strategy is composed of tree topology, replication numberdynamic optimization model and farthest node selection strategy. The farthest node isgot by farthest leaf node algorithm. Localization scheduling algorithm with waitinglength threshold is proposed in the respect of request scheduling algorithm. And theresponse is fed back to users through the HTTP protocol. Finally, it is applied totourism information retrieval system, and analyzed from the application effect. Theresults show that the data storage strategy not only can save storage space, but alsocan improve file system reliability, localization scheduling algorithm greatly increasesthe data localization probability and reduces the response time, and the response can easily feed back to user by HTTP protocol.