Accelerating R data analytics in IoT edge systems by memory optimization

De Yin Liou, Chien Chih Chen, Tien-Fu Chen, Tay Jyi Lin

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

Abstract

With the number and variety of IoT devices increasing, a large amount of data have to be moved to the cloud, resulting into unpredictable latency from limitation of network bandwidth. R language, the most popular analytic tool, has a serious bottleneck, memory garbage collection, which will become even worse problem at the edge systems with limited memory resources. Processing a large amount of dynamic objects and high percentage of LLC misses with significantmiss penalty are the two key issues in the R execution environment. In this paper, we propose a Partially Parallel Garbage Collection to improve the waiting time during garbage collection; and a centralized memory allocation mechanism to reduce miss penalty that is caused by the high percentage of LLC misses. Our optimizations can bring benefits to those machine learning algorithms that spend most of time in R for processing large data list at the edge systems.

Original languageEnglish
Title of host publication11th IEEE International Conference on Application of Information and Communication Technologies, AICT 2017 - Proceedings
PublisherInstitute of Electrical and Electronics Engineers Inc.
ISBN (Electronic)9781538605011
DOIs
StatePublished - 10 Apr 2019
Event11th IEEE International Conference on Application of Information and Communication Technologies, AICT 2017 - Moscow, Russian Federation
Duration: 20 Sep 201722 Sep 2017

Publication series

Name11th IEEE International Conference on Application of Information and Communication Technologies, AICT 2017 - Proceedings

Conference

Conference11th IEEE International Conference on Application of Information and Communication Technologies, AICT 2017
CountryRussian Federation
CityMoscow
Period20/09/1722/09/17

Keywords

  • Big data at edge
  • Data analytics
  • R language

Fingerprint Dive into the research topics of 'Accelerating R data analytics in IoT edge systems by memory optimization'. Together they form a unique fingerprint.

Cite this