A Workload-Based Dynamic Adaptive Data Replica Placement Method

Abstract

The data placement issues in cloud computing platform have been extensively researched, mainly in the choice of the number of data replica, initial data placement strategy, the run-time and dynamically adjustment and routing algorithm and other aspects of the transaction request. In this paper, we design a whole framework to the method of placing data replica for analysis and description from a higher level. We then briefly describe a method based on workload, and propose a cloud computing data replica dynamic scheduling mechanism, while in our main contribution propose a set of novel dynamic adaptive data replica placement techniques to achieve higher scalability, fault tolerance and increased variation and ability to cope with the workload. Experimental results show that our approach in data replica placement can significantly reduce the frequency of distributed transactions.