User-Information Processing Of The Cluster System Based On MPI

Based on the MPI (Message Passing Interface) parallel programming technology under the Lenovo ShenTeng cluster system platform, this thesis develops and implements a user-information processing system used to monitor and manage the situation of user resources-using and user logging information of the computing nodes in the cluster system.This thesis mainly consists of three parts: the first part introduces the related contents of the MPI parallel programming, which is the foundation of parallel inquiring operation in the cluster system. This part includes the basic concepts of parallel computing and parallel programming, introduction to MPI and MPI parallel computing environment, and also an example of matrix multiplication implemented in parallel algorithm which is compared to the traditional serial algorithm later.The second part implements the processing of user resources-using and user logging information on a single node, which is the fundamental operation of arbitrary nodes in the cluster system. The implementation is divided into three modules: user information collection and processing module, information storage module and the display module. The information collection and processing are accomplished through some core commands provided by the Linux operation system which the cluster system is running on; the information storage is accomplished through the operation of MySQL database on the server; finally, the stored information display is accomplished through JSP programming technology as a form of webpage on the client.The third part implements the user information inquiry on arbitrary nodes in the cluster system by combining the above two parts. Finally, the testing process and manipulation explanation of the whole user-information processing system are provided.This thesis mainly describes the development background, practical significance, design scheme, implementation of technology and process and the eventual process of testing and usage of the user-information processing system.