Conferences and supporting programme
End-to-End Streaming Data Management Using Streaming Analytics Framework and Visual Data Management System
With over 3 exabytes of video data expected to be flowing over the Internet by 2025, video analytics is becoming an increasingly important source of actionable information. While today's solutions have made advances in converting unstructured video data into structured information, they are often based on fixed functions and limited in their functionality. Additionally, scaling of such solutions on existing network infrastructure can be a daunting task, which mandates multidisciplinary expertise spanning across computer vision, storage, networked communications, etc. In this paper, using retail analytics as an example, we demonstrate end-to-end streaming data management using Intel's open source processing and storage frameworks ? Streaming Analytics Framework (SAF) and Visual Data Management System (VDMS). The ample experiments show that E2E streaming data management is feasible with real-time metadata acquisition and management between edge and cloud, and edge and edge, with quantitative evaluation over hundreds of devices.
--- Date: 27.02.2019 Time: 3:00 PM - 3:30 PM Location: Conference Counter NCC Ost