Spatio-Temporal Stream Processing in Microsoft StreamInsight

Mohamed Ali, Badrish Chandramouli, Balan Sethu Raman, and Ed Katibah

Abstract

Microsoft StreamInsight is a platform for developing and deploying streaming applications. StreamInsight embraces a temporal stream model to unify and further enrich query language features, handle imperfections in event delivery and define consistency guarantees on the output. With its extensibility framework, StreamInsight enables developers to integrate their domain expertise within the query pipeline as user defined functions, operators and aggregates. Also, the Microsoft SQL Server Spatial Library delivers comprehensive spatial support that enables organizations to seamlessly consume, use, and extend location-based data. This paper covers two approaches to support spatio-temporal stream processing in StreamInsight. First, the paper describes the ongoing effort at Microsoft SQL Server to bring together the temporal aspect of StreamInsight and the spatial support of the SQL Spatial Library, through the extensibility framework, to deliver an end-to-end solution for location-aware and geospatial data streaming applications. Second, the paper provides the future vision for supporting spatial attributes natively within the pipeline of the stream query processor.

Details

Publication typeInproceedings
Published inIEEE Data Engineering Bulletin
PublisherIEEE
> Publications > Spatio-Temporal Stream Processing in Microsoft StreamInsight