Apache Drill

Overview

Apache Drill is a free, open-source SQL engine for working across data stores without first centralizing their contents. It is best suited to technical teams that can manage their own deployment and want to explore or combine raw data in place. Its flexibility is a strong fit for fragmented data, but older Hadoop and Java environments may rule it out.

Key features

Drill queries raw data without requiring a prior load, schema definition or transformation. That makes it useful for exploratory queries across existing systems, especially when setting up a separate warehouse would add unnecessary steps. It does not replace data preparation where a downstream workflow depends on prepared data.

Its JSON model can handle complex and nested structures. Supported sources include HBase, MongoDB, HDFS, Amazon S3, Azure Blob Storage, Google Cloud Storage, Swift, NAS and local files, and one query can join data across multiple datastores. That combination is valuable when information is scattered across systems, though the source mix and deployment environment need to suit the organization.

JDBC and ODBC drivers connect BI tools such as Tableau, Qlik, MicroStrategy, Spotfire, SAS and Excel. Developers can also use the REST API from custom applications. Columnar execution, runtime query compilation and recompilation, locality-aware execution and a cost-based optimizer that can push processing into datastores give technical teams tools to query distributed data efficiently; the range of execution choices also makes Drill more appropriate for teams comfortable operating a query engine than for users seeking a packaged analysis app.

Drill can run embedded on a laptop or in distributed mode across a server cluster. Documented security controls include Kerberos, Plain or custom username-and-password authentication, Digest authentication, authorization, impersonation and encryption options for client-to-drillbit connections. The project is licensed under Apache License 2.0.

Pricing

Apache Drill costs nothing. The Apache License 2.0 software is downloadable and can be deployed embedded or on a cluster. It is a straightforward fit for teams that want open-source SQL querying and API access without a software charge; they should still assess the operational work of running it themselves. There is no free trial because the software is free.

Platforms

Drill supports API access and self-hosted deployment on Linux, macOS and Windows. The choice between an embedded laptop setup and a server cluster lets teams use it at different scales, while keeping deployment in their own environment.

Who it's for

Drill is a good match for developers and data teams that need SQL access to heterogeneous or nested data where it already lives, including teams connecting familiar BI clients or building custom applications. It is less suitable for buyers who want a hosted service or a managed analytics experience rather than a self-hosted engine.

Pros and cons

• Queries data in place: avoids a required load, schema-creation or transformation step for exploratory work.
• Broad source and client reach: joins across multiple datastores and connects to named BI tools through JDBC and ODBC.
• Flexible deployment: runs embedded or across a cluster, with documented authentication and authorization options.
• Compatibility caveat: official support for Hadoop 2 and Java 8 was dropped with Drill 1.22.0, so older stacks need particular scrutiny.
• Self-hosted responsibility: teams must operate the deployment themselves rather than rely on a hosted service.

Alternatives

Query Engine Software, Data Virtualization Software, Data Federation Software and SQL Query Tools offer broader category comparisons.

Choose Comunica instead if an MIT-licensed open-source option across API, self-hosted and web platforms better fits your deployment. Apache Hive is another free, open-source option for teams considering a data warehouse. Denodo Platform may suit readers who want a free developer plan with a stated single-server, four-core limit, 50 data products and 2.5 TB/year allowance, plus a trial. PrestoDB is a free, self-hosted open-source SQL query engine to consider instead.

Starburst Galaxy is worth considering when a hosted web option matters: its free plan is billed free forever and permits up to three clusters for standard ad hoc execution. Apache Arrow DataFusion suits those looking for a Rust library and CLI distributed as source artifacts. Apache Impala is another free Apache License 2.0 project with source and binary releases. Databricks Notebooks is a web-based alternative with a free edition limited to one serverless workspace and limited compute size and usage.

Verdict

Choose Apache Drill if your team needs free, self-hosted SQL queries across varied data stores and values querying raw or nested data in place. Its cross-source joins, BI drivers and REST API make it flexible for technical use. Look elsewhere if you need managed hosting, or first verify compatibility if your environment depends on Hadoop 2 or Java 8.