Towards Collaborative Data Reduction in Stream-Processing Systems

Free registration required

Executive Summary

Recent years have seen data-intensive applications that feed on near-real time 'Context' information, such as location, environmental status and surrounding resources, collected from distributed data sources leveraging sensor networks. Those data sources, such as click-streams, stock quotes and sensor data, are often characterized as fast-rate high-volume 'Streams' (Babcock et al., 2002a). Distributed stream-processing systems acquire and aggregate high-resolution data for monitoring applications that come from many different domains; these applications range from RFID-based inventory management, pipeline monitoring for civil engineering, real-time stock-price analysis, mining of web click-streams, habitat monitoring, to vital-sign monitoring and medical triage.

  • Format: PDF
  • Size: 719.8 KB