Towards a big data architecture for heterogeneous data sources

EasyChair Preprint 2747

Abstract

The use of freely available online data is rapidly increasing, as companies have detected the possibilities and the value of these data in their activities. In particular, data are seen as interesting and heterogeneous as they can, when properly treated, assist in achieving user insight into activities decision making. However, the unstructured and uncertain nature of this kind of big data presents a new kind of challenge: is there any standard architecture for big data systems?

This paper contributes to addressing this challenge by introducing a comparative architectural study to end with a unified architecture that manage data in each processing phase of the big data pipeline.

Keyphrases: Architecture, Big Data, comparative study

@booklet{EasyChair:2747,
  author    = {Latifa Rassam and Ahmed Zellou and Taoufiq Rachad},
  title     = {Towards a big data architecture for heterogeneous data sources},
  howpublished = {EasyChair Preprint 2747},
  year      = {EasyChair, 2020}}