Timestamp Embedding Query Stream Processing Engine

Abstract

Background/Objectives: To improve the performance of the streaming data through RDF vocabulary and multi query optimization. Methods/Statistical Analysis: The scalability and performance of the query processor is improved with the use of native timestamp embedding query processor with the help of semantic approach. This offers vibrant, scalable services for continuous query processing. Continuous data is controlled through dynamic sliding window for incremental evaluation of streaming data. Static queries are processed continuously and all processing is done within main memory. Semantic sensor web is used to processor sensor data semantically by query processing engine and metadata will be stored as triple stores. Findings: The continuous stream of queries is properly handled by the dynamic sliding window and sent to the query optimizer for optimization. The triple formatted outcomes are warehoused into OWL, which stores and adopt the relational database format. The data conversion will be done by the data transformer and finally results will be realized as stream of data. In this proposed model, the RDF vocabulary has been redesigned by adding timestamps along with incoming triples. The classes and properties are depicted in terms of the triples (Subject S, Object O and Predicate P) and are designed as OWL. Data sets are split into smaller parts for continuous updates and collect them into massive data sets so as to obtain faster insertion rates. It takes less time to process temporal, complex and multi queries in proposed system. Since timestamp is attached as one of the object in the triples, it is easy to retrieve and execute timestamp embedding queries. The proposed system is appropriate for all types of queries and proves efficiency in execution time. Application/Improvement: Temporal, complex and multi join queries are executed and evaluated for weather monitoring system.