With the advent of the Big Data explosion due to the Information Technology (IT) revolution during the last few decades, the need for processing and analyzing the data at low cost in minimum time has become immensely challenging. The field of Big Data analytics is driven by the demand to process Machine Learning (ML) data, real-time streaming data, and graphics processing. The most efficient solutions to Big Data analysis in a distributed environment are Hadoop and Spark administered by Apache, both these solutions are open-source data management frameworks and they allow to distribute and compute the large datasets across multiple clusters of computing nodes. This paper provides a comprehensive comparison between Apache Hadoop & Apache Spark in terms of efficiency, scalability, security, cost-effectiveness, and other parameters. It describes primary components of Hadoop and Spark frameworks to compare their performance. The major conclusion is that Spark is better in terms of scalability and speed for real-time streaming applications; whereas, Hadoop is more viable for applications dealing with bigger datasets. This case study evaluates the performance of various components of Hadoop-such, MapReduce, and Hadoop Distributed File System (HDFS) by applying it to the well-known Word Count algorithm to ascertain its efficacy in terms of storage and computational time. Subsequently, it also provides an analysis of how Spark's in-line memory processing could reduce the computational time of the Word Count Algorithm.