A datalog parallel increment maintaining method based on hierarchical counting

Abstract

The invention provides a datalog parallel increment maintaining method based on hierarchical counting, and belongs to the field of knowledge bases. The method comprises the following steps of firstly,layering an instantiation process of datalog, mapping an iterative datalog calculation process into a designed hierarchical structure, and counting each deduced fact in the hierarchical structure incombination with a counting method. In this way, the recursive datalog calculation is converted into a plurality of non-recursive calculations which are connected in series, the defect that a countingmethod cannot be applied to a recursive program is overcome, and the efficient characteristic of the counting method is reserved. The increment triggered by the updating of the epitaxial database ismaintained, and sequentially each level in the process of a counting method is updated. On the basis, an epitaxial data state sequence is further introduced, a small amount of space is utilized, and the different versions of access to an epitaxial database are provided for parallel operation. The method can process a plurality of updating operations at the same time, and greatly improves the instantiation efficiency of the epitaxial database with frequent dynamic changes.

Description

technical field [0001] The invention discloses a datalog (declarative logic programming language) parallel incremental maintenance method based on hierarchical counting, which belongs to the technical field of knowledge bases. Background technique [0002] Datalog is a declarative language similar to SQL (Structured Query Language) language. Developers only need to care about "what to do" and not "how to do it". This language allows developers to focus more on the task itself without having to consider how to achieve it. And because Datalog has more recursive semantics than SQL and has stronger expressive power, this type of language is widely used in data analysis applications, such as web page ranking, social network computing, and semantic web analysis. In order to improve the query efficiency of datalog, based on the given extension database and datalog rule program, pre-instantiation of the extension database is a common technique. At the same time, the update of the ...

AI Technical Summary

Problems solved by technology

[0004] In order to overcome the shortcomings of the traditional incremental maintenance method that cannot be parallelized when dealing with frequent update operations, the present invention provides a datalog parallel incremental maintenance method based on hierarchical counting, which can be based on the incomplete internal extension database. The update information of the extension database can be updated correctly in the extension database, so that the update operation can be processed in parallel, thereby improving the efficiency of incremental maintenance of the extension database

Images