Unbalanced Big Data-Compatible Cloud Storage Method Based on Redundancy Elimination Technology

In order to meet the requirements of users in terms of speed, capacity, storage efficiency, and security, with the goal of improving data redundancy and reducing data storage space, an unbalanced big data compatible cloud storage method based on redundancy elimination technology is proposed. A new big data acquisition platform is designed based on Hadoop and NoSQL technologies. Through this platform, efficient unbalanced data acquisition is realized. The collected data are classified and processed by classifier. The classified unbalanced big data are compressed by Huffman algorithm, and the data security is improved by data encryption. Based on the data processing results, the big data redundancy processing is carried out by using the data deduplication algorithm. The cloud platform is designed to store redundant data in the cloud. The results show that the method in this paper has high data deduplication rate and data deduplication speed rate and low data storage space and effectively reduces the burden of data storage.

On the Transformation Optimization for Stencil Computation

Stencil computation optimizations have been investigated quite a lot, and various approaches have been proposed. Loop transformation is a vital kind of optimization in modern production compilers and has proved successful employment within compilers. In this paper, we combine the two aspects to study the potential benefits some common transformation recipes may have for stencils. The recipes consist of loop unrolling, loop fusion, address precalculation, redundancy elimination, instruction reordering, load balance, and a forward and backward update algorithm named semi-stencil. Experimental evaluations of diverse stencil kernels, including 1D, 2D, and 3D computation patterns, on two typical ARM and Intel platforms, demonstrate the respective effects of the transformation recipes. An average speedup of 1.65× is obtained, and the best is 1.88× for the single transformation recipes we analyze. The compound recipes demonstrate a maximum speedup of 1.92×.