Large scale publishing and subscribing pipelined matching method based on noumenon

Abstract

The invention relates to a matching method for body-based large-scale publish-subscribe pipelines, aiming at solving the problems in the existing matching method which fails to meet the performance requirements of large-scale publish-subscribe middleware systems. The method comprises the following steps of: first establishing an RDF event graph model and an RDF subscribe graph mode; taking each arc in the RDF event graph model and the RDF subscribe graph mode as a basic semantic matching unit to establish a subscribe sentence mode index; and dividing the matching process of the basic semantic units of the RDF event graph model and the RDF subscribe graph mode into six active processes of the pipeline to form matching pipelines. The six active processes comprises reading in of typed sentences; constraining and matching of types; constraining and matching of predication; node mapping; status checking; and matched results outputting. The matching method for the body-based large-scale publish-subscribe pipelines improves the matching efficiency of the body-based large-scale publish-subscribe middleware systems, and the performance thereof is not considerably affected by the subscribed number of the systems. Simultaneously, the matching method also eliminates unnecessary and redundant matches between different subscribe graph modes.

Description

technical field [0001] The invention belongs to the technical field of computers and relates to an ontology-based large-scale publish-subscribe pipeline matching method. This method introduces ontology technology and parallel computing technology into the publish-subscribe middleware system to improve the matching accuracy and time efficiency of the large-scale publish-subscribe middleware system. Background technique [0002] The publish-subscribe middleware system is very suitable for the loose communication requirements of distributed heterogeneous platforms such as Internet large-scale information distribution, mobile computing, and grid computing, and has broad application prospects. Traditional publish-subscribe middleware systems are subject-based, content-based, XML-based, etc. Most of them rely on specific event types and simple matching mechanisms, such as: keyword matching, predicate comparison of attribute values, XPath tree pattern matching Wait. The ontology-...

AI Technical Summary

Problems solved by technology

Obviously, this method is only suitable for the case where the number of subscriptions is small or the number of variables in the subscription is limited, because when each RDF subscription graph schema contains many variables, due to the diversity and difference of variable identification and constraint conditions, it is difficult to merge these RDF Subscribing to graph patterns is very difficult and time consuming

In summary, the existing matching methods are far from meeting the performance requirements of large-scale publish-subscribe middleware systems

Images