Distributed query engine system and method and metadata server

Abstract

The invention provides a distributed key-value query engine system and a query method. The system comprises a metadata server, at least one agency node and at least one storage node, wherein the metadata server is used for maintaining the global routing list information pointing to the storage node; the agency node comprises a query forwarding module and a data forwarding module, wherein the query forwarding module is used for responding to a request which is emitted by a client end and comprises a target KEY, and forwarding the request of the client end to a corresponding storage node on the basis of the global routing list information; the data forwarding module is used for transmitting a respond package which is returned by the corresponding storage node and comprises a target Value to the client end; the storage node is used for obtaining required data from local storage on the basis of the received request of the client end, and sending the respond package comprising the target Value to the agency node. According to the distributed key-value query engine system and the query method, the key-value data distributed storage and query are realized, the massive key-value data can be contained, and the adaptability to high concurrent access is stronger.

Description

[0001] The patent application of the present invention is a divisional application of the Chinese invention patent application with the filing date of December 31, 2011, the application number of 201110460494.8, and the title of "a distributed key-value query method and query engine system". technical field [0002] The present application relates to the technical field of distributed data storage and query, in particular to a distributed key-value query method and a query engine system. Background technique [0003] For the support of large-scale Internet applications and cloud computing, relational databases are generally used to store related data. [0004] Traditional relational databases usually store a certain type of information of an application (such as website user information) in a single database table of the database system, and provide a communication interface for reading and writing operations to the application. In terms of implementation, typical relational...

AI Technical Summary

Problems solved by technology

There are two problems here. One is that for each application, the application layer needs to split the data, and the splitting strategy of each application will be different, which is inefficient; the other problem is that the cost of data re-splitting after adding machines is very high High, requires manual downtime for maintenance operations, does not support automatic elastic expansion

[0007] Secondly, in traditional relational databases, the concurrency capability of a single machine has a certain upper limit, and high-concurrency access will exhaust machine resources.

One of the reasons why high concurrent access cannot be supported is that relational databases need to perform SQL parsing operations during retrieval, which consumes a lot of CPU resources.

Images