A Survey on an Optimum Load Balancing for Large Data on Cloud Using Machine Learning

Bulky Data denotes to the huge volumes of data gathered since ages which is tedious to examine and handle using common database management tool. So we have to implement large data with data mining using linear programming or machine learning. However user can perform operation at dynamically on cloud computing. These concepts not only implement data mining operation on cloud but also perform load balancing operation. With cloud data services, it is common place for data to be not only stored in the cloud, but also to be shared across multiple users. This system of Cloud storage enables loading of data in the cloud servers efficiently and allows the user to tackle with the data without any difficulty of the resources. In the previous system, the data is deposited in the cloud using active data operation with a single cloud service provider. The cost and the Quality of service provided to the user are limited as provided by CSP. In this paper, the partitioning method is anticipated for the data repository which stores data on multiple CSPs, ensuring data security by partitioning the data which avoids the local copy at the client side. This method provides and guarantees high cloud repository integrity, improved error localization and easy identication of misbehaving server. To attain this, mobile data integrity checking concept and partition allocation using linear programming is implemented to improve the performance of cloud storage.