Parquet data files, once ingested, can be easily viewed using Hive tables by creating Hive external tables using the scripts given here. This can be executed via the Hue UI (Hive Query Builder) as detailed in earlier chapters:

SET mapred.input.dir.recursive=true;

CREATE EXTERNAL TABLE customer(id int, first_name string, last_name string, dob BIGINT)
STORED AS PARQUET LOCATION '/datalake/raw/customer';

CREATE EXTERNAL TABLE address(  id string,
                               customerId string,
                               street1 string,
                               street2 string,
                               city string,
                               state string,
                               country string,
                               zipCode string
                               )
STORED AS PARQUET LOCATION '/datalake/raw/address';

CREATE EXTERNAL TABLE contact(id string, cell string, phone string, email string)
STORED AS PARQUET LOCATION '/datalake/raw/contact';

The tables created using the preceding scripts are as shown in the following figure. The data shown is being sourced from the Parquet files:

Figure 35: Parquet Backed Hive Tables