@article{2378,
  author = {Allae Erraissi, Abdessamad Belangour, Abderrahim Tragha},
  title = {A Comparative Study of Hadoop-based Big Data Architectures},
  journal = {International Journal of Web Applications},
  year = {2017},
  volume = {9},
  number = {4},
  doi = {},
  url = {http://www.dline.info/ijwa/fulltext/v9n4/ijwav9n4_2.pdf},
  abstract = {Big Data is a concept popularized in recent years to reflect the fact that organizations are confronted with large
volumes of data to be processed and this, of course, presents a strong commercial and marketing challenge. This trend around
the analysis and collection of Big Data has given rise to new solutions that combine traditional data warehouse technologies
with Big Data systems in a logical architecture. It is important to note here that several distributions ready to use for
managing a system Big Data are available in the market, namely HortonWorks, Cloudera, MapR, IBM Infosphere BigInsights,
Pivotal HD, Microsoft HD Insight, etc. The different distributions have an approach and different positioning in relation to
the vision of a platform Hadoop. In this article, we shall first explain the architectures and components of the five distributions
of Hadoop solutions for Big Data. Then we shall present our comparative study in which we shall use 34 relevant criteria to
define the strengths and weaknesses of the main Hadoop distribution providers.},
}