Hengam: a ...

Abstract

When working with a high volume of information that follows an exponential pattern, the authors confront big data. This huge amount of information makes big data retrieval and analytics important issues. There have been many attempts to solve data analytic problems using distributed platforms, but the main problem with the proposed methods is not observing the data locality. In this article, a MapReduce-based method called Hengam is proposed. In this method, data format unification helps nodes to have data independence. The unified format leads to an increase in the information retrieval speed and prevents data exchange betoen nodes. The proposed method was evaluated using data items from an ICT company and the information retrieval time was much better than that of other open-source distributed data warehouse software.