Portal de Periódicos da CAPES

Stream processing in IoT: foundations, state-of-the-art, and future directions

2016; Springer Science+Business Media; Linguagem: Inglês

10.1016/b978-0-12-805395-9.00008-3

2199-1081

Xiao Liu, Amir Vahid Dastjerdi, Rajkumar Buyya,

Time Series Analysis and Forecasting

The emerging stream-processing paradigm has become an enabling technology that powers the time-critical applications for Internet of Things. It offers the ability to collect, integrate, analyze, and visualize continuous data streams in real time, with a scalable, highly available, and fault-tolerant architecture. This chapter outlines the relationship between stream processing and IoT, provides a discussion on the idea and formal concepts of stream processing, and further categorizes the various use-cases of the stream model into the Data Stream Management and Complex Event Processing domains. We also present a detailed analysis on the characteristics of IoT stream data with a focus on their relevant processing requirements, based on what an abstract architecture of the stream-processing system is proposed to be, which is to generally satisfy the processing needs coming from both IoT domains. This chapter concludes with an outlook that explores the future directions and trending topics regarding the development of stream processing to better support IoT applications.