Delay sensitive distributed sensor data exchange for an IoT

  • Authors:
  • Ran Tao;Stefan Poslad

  • Affiliations:
  • University of London, London, United Kingdom;University of London, London, United Kingdom

  • Venue:
  • Proceedings of the International Workshop on Adaptive Security
  • Year:
  • 2013

Quantified Score

Hi-index 0.00

Visualization

Abstract

Publish Subscribe Message Oriented Middleware or PSMOM is widely deployed in the IoT. This uses message brokers to enable open and distributed sensor data publishers (or pub) and sensor data subscribers (or sub) to exchange information asynchronously using loosely-coupled pub-sub mechanisms. Existing PSMOM messaging systems ignore the different delay requirements of different types of sensor data and thus may introduce unexpected delays to time critical sensor data, which results in a delay for critical decision making. To address this challenge, a delay sensitive aware workload management framework for PSMOM is proposed. The framework extends an existing state-of-the-art messaging system, PEER, by adding support for workload allocation, a Queue Depth load metric, and dynamic load thresholds, enabling end-to-end latency guarantees and avoiding unnecessary load balancing. The validation using a simulation is also presented in the paper.