The arrival of Big Data in society has prompted business and government to take actions to
exploit its value and application. This paper described the characteristics of Big Data and
presented an architecture for Big Data analytics. Big Data technology deviates from traditional
data management SQL-based RDBMS approaches as it deals with data with high volume,
velocity and variety. The new paradigm moves towards NoSQL databases, massively parallel
and scalable computing platforms, open-source software, and commodity servers. This paper
discussed an architecture using real-time NoSQL databases, the Hadoop HDFS distributed data
storage and MapReduce distributed data processing over a cluster of commodity servers. It
discussed running batched and real-time analytics on the Hadoop platform. Its bidirectional
relationship with traditional data warehouse and data mining analytics platform was described.
The practical significance of the architecture is to provide a blueprint for organizations moving
towards the implementation of Big Data in their enterprise. The capability for organizations to
collect and process Big Data about individuals or groups causes various privacy concerns.
Further study could address the ethical issues that may arise from Big Data and the measures that
society can take to mitigate such concerns.