Data Sources and File Formats Introduction

Included in the data sources that Drill supports are these key data sources:

  • HBase
  • Hive
  • MapR-DB
  • File system

Drill supports the following input formats for data:

  • Avro
  • CSV (Comma-Separated-Values)
  • TSV (Tab-Separated-Values)
  • PSV (Pipe-Separated-Values)
  • Parquet
  • MapR-DB*
  • Hadoop Sequence Files

* Only available when you install Drill on a cluster using the mapr-drill package.

You set the input format for data coming from data sources to Drill in the workspace portion of the storage plugin definition. The default input format in Drill is Parquet.

You change one of the store property in the sys.options table to set the output format of Drill data. The default storage format for Drill CREATE TABLE AS (CTAS) statements is Parquet.

← Data Sources and File FormatsHive-to-Drill Data Type Mapping →