Skip to content

Investigating the trade-offs of low latency responses over quality when applying machine learning algorithms over lambda architecture.

Notifications You must be signed in to change notification settings

nickozoulis/thunderstorm

Repository files navigation

thunderstorm

University of Glasgow - MSc Team Project - Fall 2015

Implemented a proof-of-concept system for realtime, accurate querying over Big Data. Our contribution was to investigate the trade-offs of low latency responses over quality when applying machine learning algorithms over Lambda Architecture.

Basic Components
  • Batch layer: Apache Spark
  • Serving layer: Apache HBase
  • Speed layer: Apache Storm

For more information build and read /report/report.tex

About

Investigating the trade-offs of low latency responses over quality when applying machine learning algorithms over lambda architecture.

Topics

Resources

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published

Languages