The Investigation on the Cost Optimization and Privacy Preservation for the Big Data Processing in Geo distributed Data center

Abstract

Sensitive Data Handling in Geo Distributed data centers concerning newlinecost is a key challenge. Many technologies being formed, managing a huge newlinecapacity of data sets in among geo-located data centers in an unsuitable newlinemanner is still boundless to provide privacy for sensitive data, resource newlineallocation, and computation time. To promise increase the efficiency, with a newlineMultivariate Metaphor based Metaheuristic Glowworm Swarm Map-Reduce newlineOptimization (MM-MGSMO) technique to address the cost reduction in newlineGDDC with flowing factors such as the impact of data allocation and an newlineoverall computation cost in the GDDC were proposed, newlineK-means algorithm is one of the outdated algorithms for the newlinegathering of data to find the centroid. The identified the centroids assign the newlinedata in each shortest data center similarly the ant colony optimization to newlinelocate the nearest optimized data center searching for the Data Manipulation newlineprocess initially one node will find the nearest optimized a path for the newlinecomputation process and it updates the colony updates another node. Data newlinecentre both algorithms were too big data analytics is difficult and random newlinedecision is a dependable one, for each iteration of the distribution of vertex is newlinepossible by probability manner to address the finding the nearest optimal newlinecentroid with hybrid framework semantic k-means Ant Colony Optimization newline(ACO) Algorithm was proposed with factors such as precision phase, Recall newlinephase and F-Measure evaluated in the GDDC. As a result, a optimistic nearest newlinecentroid solution with minimized computation cost. newline