Data access method and device adapted to super-large scale column-storage database

The invention aims to provide a rapid positioning and data access method for a super-large scale column-storage database. The data access method includes the steps that column-storage data are divided into data blocks according to features, and a multi-order feature index is established for all the data blocks. The data of the same kind are aggregated through feature division of the data, the internal dispersion of the data blocks is reduced, and access is more accurate. Filtering of the data can be accelerated through a high-order index, and access to the low-order index and data is effectively controlled. Through the two modes, data exchanging between a memory and I/O equipment will be greatly reduced, and the operation efficiency of the database will be improved.