首页--工业技术论文--自动化技术、计算机技术论文--计算技术、计算机技术论文--计算机软件论文--程序设计、软件工程论文--程序设计论文

基于Hadoop/HBase集群的在线分析处理反应调度

Abstract第7-9页
摘要第10-17页
1 INTRODUCTION第17-21页
    1.1 Online Analytical:From Centralized System to Distributed System第17-18页
        1.1.1 Motivations第17页
        1.1.2 Elements of explored solutions第17-18页
    1.2 Issues第18-19页
        1.2.1 Deploy multidimensional data over a cluster第18页
        1.2.2 Query a warehouse based on an HBase cluster第18-19页
    1.3 Contributions第19-20页
    1.4 Structure of the thesis第20-21页
2 STATE OF ART第21-34页
    2.1 Data Warehouse and OLAP第21-28页
        2.1.1 Foundations第21页
        2.1.2 Multidimensional model第21-24页
        2.1.3 Functional architecture of an OLAP system第24-25页
        2.1.4 Storage models第25-28页
    2.2 Hadoop Ecosystem第28-32页
        2.2.1 Hadoop Framework第28-29页
        2.2.2 MapReduce第29-30页
        2.2.3 HDFS:The Hadoop Dist ributed File System第30-31页
        2.2.4 HBASE第31-32页
    2.3 Data warehouse in distributed environment第32-33页
        2.3.1 Fragmentation of Warehouse第32-33页
        2.3.2 Warehouse on dist ributed database第33页
    2.4 Conclusion第33-34页
3 Multidimensional Data on Distributed Storage第34-44页
    3.1 Use Cases第34-35页
    3.2 Conceptual model for multidimensional data第35-38页
        3.2.1 Schema and Instance of Dimension第36-37页
        3.2.2 Facts and Aggregates第37-38页
        3.2.3 Local Instances of Dimension第38页
    3.3. Identification of multidimensional data第38-40页
        3.3.1 Definition and identification of multidimensional chunks第38-39页
        3.3.2 Construction of chunks blocks第39-40页
    3.4 Multidimensional data indexing第40-43页
        3.4.1 Indexes on different aggregation levels第40-42页
        3.4.2 Indexes on chunks block第42页
        3.4.3 CCB Index Operations第42-43页
    3.5 Conclusion第43-44页
4 REACTIVE SCHEDULING POLICY第44-53页
    4.1 Presentation of query processing phases第44-45页
    4.2 Rewriting the client request第45页
    4.3 Location useful data for the query第45-46页
    4.4. Queries Scheduling第46-47页
    4.5 Execution plan and optimization of execution第47页
    4.6 Queries execution and tasks scheduling第47-52页
        4.6.1 Our Scheduling Policy第48-50页
        4.6.2 Monitoring and updating the status of the execution第50-51页
        4.6.3 Assembly of the result第51页
        4.6.4 Scheduling Implementation第51-52页
    4.7 Conclusion第52-53页
5 PROTOTYPE AND EXPERIMENTATION第53-64页
    5.1 Prototype Architecture第53-57页
        5.1.1. Our data model based on HBase第53-55页
        5.1.2 Presentation of the scheduling engine services for distributed storage第55-57页
    5.2 Prototype implementation第57-60页
        5.2.1 Hadoop/HBase deployment第57-58页
        5.2.2 OLAP Client Interface第58-59页
        5.2.3 Experiments Infrastructure第59-60页
    5.3 Experiments第60-63页
        5.3.1 Test Scenario第60页
        5.3.2 Stress Scenario第60-61页
        5.3.3 Results第61-63页
    5.4 Conclusion第63-64页
6 CONCLUSIONS AND PERSPECTIVES第64-71页
    6.1 Evaluation and contributions第64-66页
        6.1.1 ldentification and indexing of data multidimensional第64-65页
        6.1.2 Implementation and Ouery Optimization第65-66页
        6.1.3 Prototype of services第66页
    6.2 Limitation and perspectives第66-71页
        6.2.1 Management and maintenance of distributed data warehouse第66-67页
        6.2.2 Maintenance and adaption of CCB Index structures according to the change of dist ributed warehouse第67-68页
        6.2.3 Evolution and optimization of query processing method第68页
        6.2.4 Design and integration of methods by services architecture第68-71页
References第71-75页
PUBLICATION第75-76页
Acknowledgments第76页
Dedication第76页

论文共76页,点击 下载论文
上一篇:基于Web的煤场可视化图形的设计与实现
下一篇:证件图像识别技术研究与应用