Big Data Server
Distributed Hadoop cluster with MapReduce and Spark
A fully distributed Hadoop cluster set up from scratch in both non-secure and secure mode, used as a platform for batch and streaming workloads.
MapReduce. A MapReduce program written in Java, informed by reading the original MapReduce paper rather than working only from framework documentation. Applied k-means clustering to document datasets through the reduce stage.
Spark. The Spark Structured API and Spark Streaming implemented across Python, Java, and Scala to compare how the same job is expressed in each language binding.