A Survey Paper on Data Spark

Abstract

There is nice interest in exploiting the chance provided by cloud computing platforms for large-scale
analytics. Among these platforms, Apache Spark is growing in quality for machine learning and graph analytics.
Developing economical advanced analytics in Spark needs deep understanding of each the algorithmic rule at hand and
also the Spark API or system genus Apis (e.g., Spark SQL, GraphX). Our BigDatalog system addresses the matter by
providing apothegmatic declarative specification of advanced queries amenable to economic analysis. Towards this goal,
we have a tendency to propose compilation and improvement techniques that tackle the necessary drawback of with
efficiency supporting rule in Spark. We have a tendency to perform in experimental comparison with alternative state-of
the-art large-scale Datalog systems and verify the effectivity of our techniques and effectiveness of Spark in supporting
Datalog-based analytics.