Database compressing and decompressing method

Abstract

The invention discloses a database compressing method which comprises: establishing compressing method set for each type of database element group and numbering the compressing method; according to the type of the database element group to select corresponding compressing method and putting its number on the head of the compressed element group; compressing the property value with the method that corresponds to the number and putting the compressed property value in the compressed element group. The invention also discloses a database decompress method which comprises selecting the compress method number from the head of element group; decompressing the element group property value according to its compressing method; putting decompressed property value in the decompressed element group.

Description

technical field [0001] The invention relates to the field of database system and information compression, in particular to a tuple-level database compression and decompression method. Background technique [0002] In a large-scale database application system, it is often necessary to process hundreds of Gb or even Tb-level data, and these data are usually stored in large-scale disk arrays. On the one hand, the contribution of the use of large-scale disk arrays to system hardware costs cannot be underestimated; on the other hand, a large number of disk read and write operations also make input / output the main performance bottleneck of the system. If the database compression method is used to compress the stored data, the disk occupation of the data can be significantly reduced, thereby greatly reducing the hardware cost of the system, and at the same time improving the overall performance of the system. [0003] Existing database compression products mainly include Winzip, W...

AI Technical Summary

Problems solved by technology

[0004] There are two main problems in using the above compression method for database compression: one is that the above compression method is suitable for compressing large text data with a high repetition rate of substrings, but the structural data mainly stored in the database does not have this feature, so It is impossible to obtain a good compression effect; the second is that the encoding table needs to occupy a large amount of space. Taking Winzip as an example, even if a data containing only one byte is compressed, the length of the compressed data will reach more than 200 bytes, while the database The length of most of the tuples in will not exceed this length, so in order to achieve reasonable compression efficiency, the compression and decompression process must be performed in larger units, usually a single page or the entire file, which means that users want to To access a tuple, the entire page or the entire file where the tuple is located must be compressed or decompressed, which greatly increases the processing overhead

Images