作者: Yuxin Jing , Hanpin Wang , Yu Huang , Lei Zhang , Yongzhi Cao
DOI: 10.1007/S11265-017-1298-7
关键词:
摘要: MapReduce is a powerful distributed data analysis programming model. It runs on big storage systems and processes in parallel way. An appropriate way to ensure the correctness of programs formal method analysis, which requires firstly model MapReduce. In this paper we propose modeling language establish framework. Unlike other approaches, our describes processing from perspective underlying files blocks, so that details can be clearly demonstrated. The based previous work, describing management massive systems, with extensions two aspects: block content refinement concurrency support. Based language, features discussed.