Automating Characterization Deployment in Distributed Data Stream Management Systems

Authors

  • Dhanasekar.S Assistant Professor, Info Institute of Engineering, Kovilpalayam, Coimbatore- 641 107
  • Chaithanya.M.N UG graduate, Info Institute of Engineering, Kovilpalayam, Coimbatore- 641 107
  • Gayathri.R.S UG graduate, Info Institute of Engineering, Kovilpalayam, Coimbatore
  • Nandhini.N UG graduate, Info Institute of Engineering, Kovilpalayam, Coimbatore

Keywords:

RQS, SPS, Four level feature extraction, Optimal resource configuration, Candidate settings, SPS-Storm

Abstract

DDSMS composed of two layers: upper layer – Relational Query Systems (RQS) and lower layer – Stream
Processing Systems (SPS).After query submission to RQS, query planner needs to get converted into DAG consisting
tasks running on SPS.SPS configure different deployment strategies based on query requests and data stream
properties.Introducing four-level feature extraction, it uses different query workload as training sets to predict resource
usage.Select optimal resource configuration from candidate settings based on current query requests and stream
properties.Finally, validate the approach on open source SPS-Storm.

Downloads

Published

2018-03-25

How to Cite

Dhanasekar.S, Chaithanya.M.N, Gayathri.R.S, & Nandhini.N. (2018). Automating Characterization Deployment in Distributed Data Stream Management Systems. International Journal of Advance Engineering and Research Development (IJAERD), 5(3), 210–216. Retrieved from https://ijaerd.org/index.php/IJAERD/article/view/2642

Issue

Vol. 5 No. 3 (2018): Volume 5 Issue 3 March 2018

Section

Articles

Most read articles by the same author(s)