Flash-Based Database Indexing Technology Research

Flash Memory is a read-only memory which is erasable and programmable. It is a data storage device widely used in mobile electronic products with the characters of being erased and written many times. Flash Memory has many advantages, such as small volume, light weight, fast access speed and wide suitable environment, and so on. Recently, with the improvement of flash memory technology, the usability of flash memory device is greater, especially for the solid-state disk. SSD is used in large enterprise data storage system more and more widely, such as database system. The access speed of flash memory is so fast that its random access performance is much better than traditional disk. Therefore, it is just the problem of time that flash memory takes the place of magnetic disk.Index technique, as an important component for database management system, has a significant effect on the database comprehensive performance. However, current index techniques in database management system are all designed according to traditional disk characteristics. It cannot get the best performance if we just transplant the index techniques from traditional database to flash memory device. Therefore, we need to design the index structure for flash memory device specially. Based on this problem, this paper analyses current research achievement for database index technique, and proposes a CLR-Tree index method based on R-Tree. CLR-Tree combines the advantages from R-Tree and log management, and converts the random update operators to ordered update. In addition, CLR-Tree also compresses the log which accelerates the query speed of log. Finally, the environment result shows the performance of CLR-Tree is better the existing methods.