Unlock instant, AI-driven research and patent intelligence for your innovation.

Multi-data center synchronization method supporting remote pulling

A multi-data center, data synchronization technology, applied in the fields of digital data processing, structured data retrieval, special data processing applications, etc. Synchronization volume, improve synchronization efficiency, and ensure the effect of performance

Pending Publication Date: 2022-04-05
CHENGDU SOBEY DIGITAL TECH CO LTD
View PDF0 Cites 0 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

This method is flexible enough, but the synchronization granularity is too fine, and the result set faces organization and compression problems during transmission.
Cannot give full play to the advantages of time series data in orderly organization of time

Method used

the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
View more

Image

Smart Image Click on the blue labels to locate them in the text.
Viewing Examples
Smart Image
  • Multi-data center synchronization method supporting remote pulling
  • Multi-data center synchronization method supporting remote pulling

Examples

Experimental program
Comparison scheme
Effect test

Embodiment 1

[0044] Embodiment 1: A multi-data center synchronization method that supports remote fetching, comprising the steps of: when synchronizing data between multiple data centers, fetching is performed according to the data involved in the user SQL statement, and the data block is used as a unit Perform on-demand synchronization; and, when the user performs down-sampling query, only the relevant statistical information is synchronized, but not the data itself; in the process of data compression, there is no need to decode the data block, and there is no need to re-decode it after transferring it to the target database. coding.

Embodiment 2

[0045] Embodiment 2: on the basis of embodiment 1, include sub-steps:

[0046] S1, the local database receives the synchronous SQL statement, analyzes the statement, and judges whether it is necessary to pull the data remotely. If the data needs to be pulled, the statement is transmitted to the remote database, and the SQL statement is judged again to pull the corresponding data;

[0047] S2, transfer the data blocks to be synchronized to the local database;

[0048] S3, rebuilding the upper-lower relationship of the statistical information of the data block.

Embodiment 3

[0049] Embodiment 3: On the basis of Embodiment 2, in step S1, sub-steps are included: when the user executes a query in the local database, the local database will first judge whether the queried data exists in the local database, and if so, directly from the local database. The database acquires data, and if it does not exist, data synchronization is initiated to the remote database in units of Data Pages and the storage location of the data block of the corresponding layer is moved out in advance.

the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
Login to View More

PUM

No PUM Login to View More

Abstract

The invention discloses a multi-data center synchronization method supporting remote pulling, which belongs to the technical field of time sequence databases and comprises the following steps: when data synchronization is carried out among multiple data centers, pulling is carried out according to data involved in a user SQL (Structured Query Language) statement, and on-demand synchronization is carried out by taking a data block as a unit; moreover, when a user carries out downsampling query, only related statistical information is synchronized, and data is not synchronized; in the process of compressing the data, decoding is not needed when the data blocks are obtained, and re-encoding is not needed after the data blocks are transmitted to the target library. The method is particularly suitable for multi-data center synchronization under the condition that the bandwidth is limited, the synchronization does not need a special SQL statement, the synchronization operation is non-inductive to a user, and the synchronization efficiency is improved.

Description

technical field [0001] The present invention relates to the technical field of time-series databases, and more specifically, relates to a multi-data center synchronization method supporting remote fetching. Background technique [0002] If traditional databases want to synchronize data from remote to local, there are usually two ways: [0003] (1) Database cluster: In this way, the data in each database in the cluster needs to be kept in real-time synchronization, so the requirements for network bandwidth are high. In an environment with high network delay, the performance of the cluster is low, and a large number of transaction rollbacks are prone to occur. . Moreover, the synchronization method is not flexible enough. Even if a local user only cares about a small amount of remote data, the entire database must be synchronized, resulting in a waste of storage resources. [0004] (2) Remote query + local insertion: This method is to query data from a remote table and then ...

Claims

the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
Login to View More

Application Information

Patent Timeline
no application Login to View More
IPC IPC(8): G06F16/27G06F16/242
Inventor 吴春中张浩阳
Owner CHENGDU SOBEY DIGITAL TECH CO LTD