Cloud computing service system and method for massive dataset parallel computation

Abstract

The invention provides a cloud computing service system and a cloud computing service method for massive dataset parallel computation. The system comprises a cloud computing operation system CloudView, a virtual machine platform, a MapReduce computation platform, an automated configuration management tool and a client, wherein the virtual machine platform distributes virtualized hardware resources of the cloud computing operation system CloudView to a virtual machine according to client requirements; the MapReduce computation platform is established in the virtual machine; and parallel computation service is distributed to the client through the MapReduce computation platform; the client performs data interaction with the MapReduce computation platform through the virtual machine; and the automated configuration management tool adjusts parameters and configurations of the MapReduce platform according to actual needs. The cloud computing service system and the cloud computing service method for massive dataset parallel computation use a uniform management platform, resources are fully used and shared, application isolation is good, the requirements of a plurality of uses and a plurality of tasks are well met, and the cloud computing service system and the cloud computing service method are easy to manage and maintain.

Description

technical field [0001] The invention belongs to the field of cloud computing, and in particular relates to a cloud computing service system and method for parallel computing of large-scale data sets. Background technique [0002] The MapReduce computing platform is a software architecture for parallel computing of large-scale data sets. At present, more and more applications need to rely on the MapReduce platform to realize calculations, especially for the processing of massive data. MapReduce has shown its efficient processing capabilities. Due to the good scalability, fault tolerance and availability of MapReduce, various industries are currently using MapReduce computing platforms, and various scientific research institutes are also researching and using MapReduce. In short, the MapReduce computing platform is already one of the necessary tools. Since MapReduce requires computer clusters to provide efficient processing capabilities, how to effectively implement the depl...

AI Technical Summary

Problems solved by technology

[0007] The current implementation scheme of the MapReduce cloud computing platform, because there is no unified management platform, the resources are not fully utilized and shared, the application isolation is poor, and the needs of multi-users and multi-tasking cannot be well met, and it is troublesome to use. Not easy to manage and maintain, let alone provide MapReduce computing services

Images