Large scale stream analytics using a resource-constrained edge

Roshan Bharath Das, Gabriele Di Bernardo, Henri Bal

Research output: Chapter in Book / Report / Conference proceedingConference contributionAcademicpeer-review

Abstract

A key challenge for smart city analytics is fast extraction, accumulation and processing of sensor data collected from a large number of IoT devices. Edge computing has enabled processing of simple analytics, such as aggregation, geographically closer to the IoT devices to improve latency. However, the throughput of processing in the edge depends on the type of resources available, the number of IoT devices connected and the type of stream analytics performed in the edge. We introduce a framework called Seagull for building efficient, large scale IoT-based applications. Our framework distributes the stream analytics processing tasks to the nodes based on their proximity to the sensor data source as well as the amount of processing the nodes can handle. Our evaluation shows the effect of various stream analytics parameters on the maximum sustainable throughput for a resource-constrained edge device.

Original languageEnglish
Title of host publicationProceedings - 2018 IEEE International Conference on Edge Computing, EDGE 2018 - Part of the 2018 IEEE World Congress on Services
PublisherInstitute of Electrical and Electronics Engineers Inc.
Pages135-139
Number of pages5
ISBN (Electronic)9781538672389
DOIs
Publication statusPublished - 26 Sep 2018
Event2018 IEEE International Conference on Edge Computing, EDGE 2018 - San Francisco, United States
Duration: 2 Jul 20187 Jul 2018

Conference

Conference2018 IEEE International Conference on Edge Computing, EDGE 2018
CountryUnited States
CitySan Francisco
Period2/07/187/07/18

Keywords

  • Edge analytics
  • IoT framework
  • Smart city analytics
  • Stream processing

Fingerprint Dive into the research topics of 'Large scale stream analytics using a resource-constrained edge'. Together they form a unique fingerprint.

Cite this