Cluster Layout
Vertica and Hadoop each use a cluster of nodes for distributed processing. These clusters can be co-located, meaning you run both products on the same machines, or separate. See Co-Located Clusters and Separate Clusters.
With either architecture, if you are using the hdfs
scheme to read ORC or Parquet files, you must do some additional configuration. See Configuring the hdfs Scheme.