Database is the foundation of the electronic, the business intelligence and ERP system, usually saving the most important information. However, with the data quantity and the business volume fast growth, and the high concurrent visit’s appearance, cause traditional the stand-alone database unbearable, so the system’s performance and availability urgently need to be solved. As to the business intelligence, the business people need to conduct information through data analysis and data mining, converting the business data into effective data. The foundation of business intelligence is providing uninterrupted data services for business people, the research on high-availability database is imperative.In this thesis, Car business intelligence solutions puts SQL Server as database management software of background server. As the cost advantage of SQL Server compared with Oracle, more and more enterprises use it as the data management and analysis solution platform. The process of business intelligence mainly includes ETL, multidimensional data analysis and data mining. This thesis emphasizes the importance of data in business intelligence solutions through introducing the overall architecture of car business intelligence solutions, and then to study the high-availability database technology.Failover clustering technology is the most important technology in the always on high availability group. SQL Server2012adds support for multi-site clustering to achieve cross-domain sub-functions compared with the original version, this function requires two or more data centers to achieve local high availability and remote disaster recovery.Then compared the failover cluster and Oracle Real Application Clusters, as the two-node Oracle Real Application Cluster and failover cluster in the Always On for instance, the configuration and maintenance are more easier under the conditions of satisfied high availability. Finally, testing the high availability solution based on database mirror technique, it shows the high availability solution based on cluster technique is the best suitable solution for the databse structure of enterprise, and puts forward a high availability solution based on Microsoft official solution, then give the analyse of high availability test. |