作者: Tao Wang , Shihong Yao , Zhengquan Xu , Shan Jia , Qiang Xu
DOI: 10.1109/SMARTCITY.2015.139
关键词: Cloud computing 、 Computational complexity theory 、 Big data 、 Distributed data store 、 Dynamic priority scheduling 、 Computer science 、 Distributed database 、 Distributed computing 、 Fair-share scheduling 、 Data center 、 Data mining
摘要: In complex and data-intensive applications, data scheduling between centers must occur when multiple datasets stored in distributed are processed by one computation. To store massive effectively reduce during the execution of computations, a mathematical model cloud computing is built dynamic computation correlation (DCC) defined. Then placement strategy for big based on DCC proposed. Datasets with high placed into same center, new dynamically most appropriate center. Comprehensive experiments show that proposed can number has considerably low almost constant computational complexity increases massive. It be expected will applicable to practical large-scale storage systems management.