一种基于海量信息处理的云存储模型研究.pdf

计算机研究与发展
ISSN 1000- 11-1777?TP
Journal puter Research and Development 49(Suppl.):32-36,2012
一种基于海量信息处理的云存储模型研究
张桂刚李超张勇邢春晓

清华大学信息技术研究院北京
( 100084)
清华大学信息科学国家实验室北京
( 100084)
(******@.)
A Kind of Cloud Storage Model Research Based on Massive Information
Processing
Zhang Guigang,Li Chao,Zhang Yong,and Xing Chunxiao
(Research Institute of Information Technology,Tsinghua University,Beijing100084)
(Tsinghua National Laboratory for Information Science and Technology,Tsinghua University,Beijing100084)
Abstract This paper presents a kind of cloud storage model based on massive information processing.
The model includes:1)storage methods of structured,semi-structured and unstructured data and the
corresponding resources;2)THDFS (Tsing Hua Distributed File System)which will take into
account all size of the massive data files(big files and small files);3)a cloud database system design
model THCloudDB (Tsing Hua Cloud Database);4)a -base massive data processing
computation model THMapReduce(Tsing Hua putation model based on hadoop++
which is stronger than MapReduce and weaker than SQL,especially it have merged into the semantic
computing technology;5)a intelligent mining tool THBI(Tsing Hua Business Intelligent)based on
the THDFS and THCloudDB,and it have the semantic ,paperend takes“Tsinghua
Tong”as a case and make som