Chinese Journal of Computers   Full Text
Title	A Data Placement Strategy for Data-Intensive Applications in Cloud
Authors	`ZHENG Pai CUI Li-Zhen WANG Hai-Yang XU Meng
Address	(School of Computer Science and Technology, Shandong University, Jinan 250101)
Year	2010
Issue	No.8(1472—1480)
Abstract &   Background	Abstract With the development of information technology, data-intensive applications in cloud have been used in more and more fields. Because of the decentralized data centers in cloud, these applications now are facing some new challenges in data placement which mainly include how to reduce the time cost of data movements between data centers, how to deal with the data dependencies, and how to keep a relative load balancing of data centers. This paper proposes a data placement strategy, the three stages of which address the three challenges above respectively. Simulation shows that the strategy can effectively reduce the time cost of data movements across data centers during the application’s execution. Keywords cloud computing; process; data-intensive; data placement; data dependency Background This work is supported by the National Nature Science Foundation of China under grant No. 90818001, Key Technology R&D Program of Shandong Province under Grant No. 2008GG30001005 and No. 2009GG10001002, Specialized Research Fund for the Doctoral Program of Higher Education under grant No. 200804221031, and Independent Innovation Foundation of Shandong University under grant No. 2009TS030. Faced with the development of cloud computing, data placement which is a critical aspect of data management for data-intensive applications in cloud is gaining more and more attention. The primary challenge of data placement is the inevitable data movement between distributed data centers in cloud. The main objective of this paper is to provide a data placement strategy to reduce the time cost of data movement between distributed data centers while taking data dependency and load balancing into consideration. Compared with an existing data placement strategy which focuses on the reduction of the data movement during the application’s execution, our strategy can reduce the time cost of this data movement more efficiently.