How does the MapReduce component of a big data analytics solution process a large data set? Implements a high-level query layer to describe and run data processing jobs Uses the HDFS protocols for advanced caching on the NameNode of an HDFS cluster Replicates the entire data set on multiple high-performance compute nodes for processing Splits the data set into parts and distributes them across multiple computing nodes in a cluster What is a virtual storage system in a software-defined storage environment? A pool of object-based storage systems from which block and file storage are provisioned as services to users