Method for storing high-precision map data based on distributed table and related equipment thereof

Abstract

The invention discloses a method for storing high-precision map data based on a distributed table and related equipment thereof. The method comprises the steps of obtaining high-precision map data to be stored; determining a service logic requirement corresponding to the high-precision map data; establishing a corresponding data table according to the business logic requirements and the high-precision map data; and carrying out distributed storage on the data in the corresponding data table according to the business logic requirement. According to the method, the storage reliability can be provided, the data management is finer, and the corresponding storage of high-precision map data is realized according to the service logic, so that the data downloaded to a certain area of a picture sheet can be quickly indexed during data access, and the data reading efficiency is improved.

Description

technical field [0001] The present invention relates to the technical field of map data processing, in particular to a method, device, computer equipment and storage medium for storing high-precision map data based on distributed tables. Background technique [0002] High-precision map data is more refined than traditional map data elements, and the corresponding data volume will be very large. The large amount of data includes two aspects: one is the large storage capacity of the data. At present, the storage cluster capacity has reached the PB (PetaByte, gigabyte) level, and there will be several more after the national network construction and urban road production advance. Double the capacity requirements. The second is that the order of magnitude of the data bar is very large, and there are many fine-grained data elements, and a large graph can easily reach millions of elements. [0003] In related technologies, map data usually adopts the following three storage meth...

AI Technical Summary

Problems solved by technology

[0004] However, the above three storage methods have obvious disadvantages for storing such a large amount of high-precision map data: the first method, NFS is a common file sharing protocol, but its reliability is not high, and it has strict requirements on the network environment, especially It is difficult to maintain when crossing LANs. At the same time, the granularity of the entire file hosting method is too coarse, which has limitations on business logic; the second method uses PG libraries for storage, and there are particularly many fine elements of high-precision maps, resulting in relationship libraries. The entries will be very large, the index will be very large, and the performance will decrease as the amount of data increases; the third way is to use AFS, which is a distributed file storage that supports large-scale storage. All data is recorded in files, but requires A large amount of index information is recorded in the business layer in order to locate the file path where the data is located, offset (the beginning of the file) and size (the total length of the read content); in addition, multiple requests for concurrent writing cannot guarantee data security, and append (data append ) data writing method is not suitable for the expansion of business logic, resulting in complex business logic

Images