Uploading Deferrable Big Data to the Cloud by Improved Dynamic Self-Adaption Algorithm

Cloud computing is a pattern of processing the big data and provides the convenient, on-demand network access to a shared pool of configurable computing resources. Cloud data center´s cost is becoming the hot topic in recent years. 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 study the deficiency of randomized online algorithm and Dynamic Self-adaption Algorithm, and we propose the improved algorithm to minimize the bandwidth cost based on DSA, and the results demonstrate that the IDSA´ performance is better than the ROA and DSA.