An Optimized Method for Access of LOBs in Database Management Systems
暂无分享,去创建一个
This paper researches methods used by relational database management systems to manage LOBs. It attempts to use LOB slave table methods to optimize and advance the storage structures of large objects as well as optimize the redo and undo log mechanisms to increase LOB access efficiency. For changes in slave table data, it effectively prevents the redo log from being written into the LOB database, while diminishing the amount of changes being recording within the undo journal. This improves the writing speed of the log, thus improving the access efficiency and speed of the slave table record. At the same time, this optimized log mechanism also ensures the correct recovery of data after a system crash. Tests have shown that the optimized LOB access efficiency is slightly better than Oracle’s.
[1] Geoffrey C. Fox,et al. Cloud computing paradigms for pleasingly parallel biomedical applications , 2010, HPDC '10.
[2] Peter J. Haas,et al. Ricardo: integrating R and Hadoop , 2010, SIGMOD Conference.
[3] Jinyun Fang,et al. Multi-dimensional Index on Hadoop Distributed File System , 2010, 2010 IEEE Fifth International Conference on Networking, Architecture, and Storage.