Multi-path cross-class query and optimization method in object proxy database

Abstract

The invention proposes a multi-path cross-class query and an optimization method in an object proxy database, and defines a multi-path cross-class query that uses different path expressions on the same path to respectively obtain target attributes as a multi-path cross-class query with the same terminal class. For the case that multiple paths in the cross-class query contain common class nodes, it is defined as a multi-path cross-class query with different terminal classes. Next, a multi-path cross-class query syntax support implementation is proposed, which is used to support multi-path cross-class queries with the same end point class and different end point classes. Finally, the execution plan of multi-path cross-class query is proposed, and based on the idea of ​​pointer tracking algorithm used in the original cross-class query, a multi-path pointer tracking algorithm for multi-path cross-class query is proposed on this basis, which is used for multi-path expressions calculation. By adopting the present invention, the user can use one path expression to obtain different target attribute expressions, and the redundant representation of common path expressions is reduced, and the query efficiency is improved.

Description

technical field [0001] The invention belongs to the technical field of database query optimization, in particular to a multi-path cross-class query and optimization method in an object proxy database. Background technique [0002] With the rapid development of cloud computing and big data technology, database applications are constantly updated, and new application fields provide new directions for the development of database technology, and also put forward new requirements. Massive data is no longer limited to structured patterns, and semi-structured, unstructured, and other complex-structured data have emerged. The traditional relational data model mainly manages structured data, so it seems stretched to face this situation. The object-oriented database proposes a management scheme for complex data, but because it uses encapsulation to manage semi-structured and unstructured data, it makes it difficult to move, split and reorganize objects, so it loses the unique flexibi...

AI Technical Summary

Problems solved by technology

[0006] (1) When the query contains multiple path expressions with common parts, independent execution between multiple path expressions will cause a large number of repeated accesses to objects

For this kind of demand for obtaining attributes in different objects in the same object network, there is currently a lack of a query function that enables users to obtain different target attributes by writing only one path expression

[0007] (2) Currently in ODDB, since multiple path expressions need to be used to obtain target attribute values ​​independently of each other, and the "real-time calculation" mode of one object at a time is adopted, when the queried proxy class has a higher proxy level, multiple There are relatively many public class nodes between the two paths, and there are a lot of repeated access to objects in the public class when obtaining the target attribute

Repeated access to objects requires repeated scanning of the bidirectional pointer table. Frequent access will result in a large number of I / O operations, which seriously affects the efficiency of cross-class queries

Images