Massive Traffic Data Storage Method

Abstract

The invention provides a massive traffic data storage method. The massive traffic data storage method comprises the following steps: enabling obtained massive traffic data to generate into a data matrix according to an object-attribute mapping relation; segmenting the data matrix into multiple subdata matrixes according to a set rule; and carrying out fusion calculation on each of the subdata matrixes to form a reconstructed storage structure and carrying out sorted storing on the reconstructed data matrix. The massive traffic data can be stored quickly and effectively, and the storage efficiency and the finding-calling efficiency of the data are improved; the data loss in a storage process is effectively prevented, and the nonvolatility of the traffic data is improved; and thus, the multi-user write-in and modification can be supported.

Description

technical field [0001] The invention relates to a data storage method, in particular to a mass traffic data storage method. Background technique [0002] With the rapid development of the transportation industry, the traffic data has shown explosive growth, and the data structure and the relationship between the data have also become more and more complex. Therefore, how to store massive data and facilitate subsequent extraction and analysis has become an important issue. Technical problems that need to be solved urgently. [0003] In the prior art, the usual method for data storage is to store data according to the corresponding relationship between objects and attributes, which is feasible for small data storage and can meet the requirements of subsequent use, but for massive In terms of data, the existing storage method needs to establish a lot of mapping relationship tables. When calling data after data storage, it is often difficult or requires an extremely complicated...

AI Technical Summary

Problems solved by technology

[0003] In the prior art, the usual method for data storage is to store data according to the corresponding relationship between objects and attributes, which is feasible for small data storage and can meet the requirements of subsequent use, but for massive In terms of data, the existing storage method needs to establish a lot of mapping relationship tables. When calling data after data storage, it is often difficult or requires an extremely complicated algorithm to establish a one-to-one correspondence between the mapping relationship tables. In the process, it is necessary to perform associative traversal of the data matrix according to the mapping relationship, which greatly reduces the efficiency of data search and call. Moreover, the existing technology method easily leads to incomplete data during the data storage process, that is, data loss, and cannot Support multi-user writing and modification, thus affecting subsequent use

Images