apache/druid

★ 13,993JavaAudience · dataComplexity · 4/5LicenseSetup · hard

Mindmap

mindmap
  root((Apache Druid))
    Data Ingestion
      Streaming sources
      Batch files
    Querying
      SQL interface
      HTTP API
      Web console
    Architecture
      Cluster processes
      Independent scaling
    Use Cases
      Product dashboards
      Real-time monitoring
      Event analytics

mindmap root((Apache Druid)) Data Ingestion Streaming sources Batch files Querying SQL interface HTTP API Web console Architecture Cluster processes Independent scaling Use Cases Product dashboards Real-time monitoring Event analytics

Click or tap to explore — scroll the page freely

Things people build with this

USE CASE 1

Power a product analytics dashboard that must return query results instantly even with millions of daily events.

USE CASE 2

Ingest streaming event data and make it queryable within seconds for real-time monitoring.

USE CASE 3

Run ad-hoc SQL queries over years of historical log data without pre-aggregating it.

USE CASE 4

Replace a slow data warehouse for high-concurrency, time-series query workloads.

Tech stack

JavaSQLHTTP APIDocker

Getting it running

Difficulty · hard Time to first run · 1h+

Runs as a cluster of multiple separate processes, Docker quickstart simplifies local setup but production requires careful cluster sizing and configuration.

Use, modify, and distribute freely under the Apache 2.0 license, including in commercial products.

In plain English

Apache Druid is a database built for answering questions about large amounts of data very quickly. It is designed for situations where you need fast answers from data that is arriving continuously, such as tracking user activity on a website in real time or monitoring events as they happen. Think of it as a tool that sits between raw incoming data and the people or dashboards that need to query it immediately. Druid is particularly good at handling high traffic queries, like powering the charts and tables inside a product dashboard that many users might be viewing at once. It can ingest data from streaming sources (data arriving in a continuous flow) and from batch sources (large files loaded at scheduled times). Once data is loaded, queries typically return in a fraction of a second even over billions of rows. The project includes a web console where you can set up data loading, browse what data is stored, and run queries without writing code. For developers, it also exposes SQL and HTTP interfaces so applications can connect to it directly. It runs as a cluster of separate processes, meaning you can scale different parts of the system independently depending on where the bottleneck is. Druid is an open source project under the Apache Software Foundation, so it is free to use and has an active community. Official documentation, quickstart guides for running it locally or in a container environment, and community forums are all linked from the repository.

Copy-paste prompts

Prompt 1

I have a Kafka stream of user click events and want to query them in real time with sub-second latency. How do I ingest that data into Apache Druid and write a SQL query to count events per page per minute?

Prompt 2

Set up Apache Druid locally using Docker and show me how to load a sample CSV file and run my first query using the web console.

Prompt 3

My product dashboard runs slow queries against a PostgreSQL table with 500 million rows. How would I migrate that data into Druid to get millisecond query times?

Prompt 4

Explain how Apache Druid's cluster architecture works and which processes I should scale up first when queries slow down under load.

Open on GitHub → Explain another repo

← apache on gitmyhub — every repo by this author, as a profile.

Verify against the repo before relying on details.