The Delta Big Data Architecture for Mobility Analytics
George A. Vouros, Apostolis Glenis, Christos Doulkeridis · 2020
Motivated by requirements in mobility analytics that require joint exploitation of streamed and voluminous archival data from diverse and heterogeneous data sources, this paper presents the δ architecture. Denoting "difference", δ emphasises on the different processing requirements from loosely-coupled components, which serve intertwined processing purposes, forming processing pipelines. The δ architecture, contributes principles for realizing systems, focusing on the requirements from components and pipelines, which are specified as constraints on performance indicators. The article presents a specific instantiation of the δ architecture to satisfy requirements for big data mobility analytics, exploiting real-world mobility data for performing realtime and batch analysis tasks.