Apache Drill

Features

One explicitly stated design goal is that Drill is able to scale to 10,000 servers or more and to be able to process petabytes of data and trillions of records in seconds.

• Schema-free JSON document model similar to MongoDB and Elasticsearch, without requiring a formal schema to be declared
• Industry-standard APIs: ANSI SQL, ODBC/JDBC, RESTful APIs
• Extremely user and developer friendly
• Pluggable architecture enables connectivity to multiple datastores
• Version 1.9 added dynamic user-defined functions
• Version 1.11 added cryptographic-related functions and PCAP file format support

Back-end Support

Drill is primarily focused on non-relational datastores, including Apache Hadoop text files, NoSQL, and cloud storage. A notable feature also includes in situ querying of local JSON and Apache Parquet files. Some additional datastores that it supports include:

• All Hadoop distributions (HDFS API 2.3+), including Apache Hadoop, MapR, CDH and Amazon EMR
• NoSQL: MongoDB, Apache HBase, Apache Cassandra
• Online Analytical Processing: Apache Kudu, Apache Druid, OpenTSDB
• Cloud storage: Amazon S3, Google Cloud Storage, Azure Blob Storage, Swift, IBM Cloud Object Storage
• Diverse data formats, including Apache Avro, Apache Parquet and JSON
• RDBMs storage plugins (Using JDBC to connect to MySQL, PostgreSQL, and others)

A new datastore can be added by developing a storage plugin. Drill's "schema-free" JSON data model enables it to query non-relational datastores in-situ .

Front-end Support

Drill itself can be queried via JDBC, ODBC, or REST through a variety of methods and languages including Python and Java. The default install includes a web interface allowing end-users to execute ANSI SQL directly and export data tables as CSV files without any programming.

The dashboard library, Apache Superset, is particularly well suited for visualization of data queried with Drill.

References

Papers

Some papers influenced the birth and design. Here is a partial list:

• 2005 From Databases to Dataspaces: A New Abstraction for Information Management, the authors highlight the need for storage systems to accept all data formats and to provide APIs for data access that evolve based on the storage system's understanding of the data.
• 2010 Dremel: Interactive Analysis of Web-Scale Datasets

References

1. Friedman, Ellen. (21 Sep 2015). "Apache Drill: Tracking its history as an open source community".
2. "Brief About The Differences between Apache Drill Vs Presto".
3. "Spark SQL vs. Apache Drill-War of the SQL-on-Hadoop Tools".
4. (2 December 2014). "The Apache Software Foundation Announces Apache Drill as a Top-Level Project".
5. "Apache Drill - Schema-free SQL for Hadoop, NoSQL and Cloud Storage".
6. Vizard, Michael. (2021-09-01). "Apache Software Foundation updates Drill for broader SQL queries".
7. (2016-04-11). "Apache Drill Eliminates ETL, Data Transformation for MapR Database".
8. "DrillProposal - INCUBATOR - Apache Software Foundation".
9. "Frequently Asked Questions - Apache Drill".
10. Wayner, James R. Borck, Martin Heller, Steven Nuñez, Andrew C. Oliver, Ian Pointer and Peter. (2020-10-05). "The best open source software of 2020".