A Dynamic Self-Adaptive Algorithm for Uploading Deferrable Big Data to the Cloud Cost-Effectively

Cloud computing is based between the service provider and service consumer agreements and cloud data center is under a cloud computing environment that consists of hardware and software components. This paper studies how to minimize the bandwidth cost for uploading deferral big data to a cloud computing platform, based on the MapReduce Framework. We first analysis the shortcoming of bandwidth of data centers, and we provide an optimization algorithm -- Dynamic Self-adaption Algorithm, it will reduce the cost of bandwidth of data centers. Dynamic Self-adaption Algorithm optimize the total cost of network bandwidth. Compared to a random selection algorithm, Dynamic Self-adaption Algorithm efficiently in this paper used of network peak to transfer data, and reduces the probability of spare channels appear to improve bandwidth utilization, so it reduces bandwidth costs.