druid-io/druid
Druid
Druid is a distributed columnar store and online analytical processing database designed for real-time analytics. It functions as a SQL analytics platform and a streaming data ingestion engine, allowing for the analysis of large datasets with low latency to support interactive dashboards and high-concurrency operational workloads.
The system integrates a streaming data ingestion engine that loads information via batch or streaming processes to enable immediate analysis of arriving data. It provides high-performance analytical processing to execute slice-and-dice queries on massive data volumes for trend and pattern identification.
The platform includes capabilities for distributed database management and cluster monitoring through SQL system tables. It supports data retrieval via standardized query languages and web-based application programming interfaces.
Features
- Real-time Analytics Platforms - Provides a high-performance platform for interactive, low-latency analytical processing and slice-and-dice queries.
- Analytical Databases - Provides an analytical database engine tuned for high-concurrency operational workloads and complex queries.
- Columnar Databases - Functions as a distributed columnar database for efficient aggregation and retrieval of massive datasets.
- Columnar Storage Engines - Implements a columnar storage engine optimized for high-compression and fast analytical aggregation.
- Data Ingestion - Loads data through coordinated streaming and batch pipelines using specialized supervisors.
- Lambda Architectures - Integrates a lambda-style pipeline combining real-time streaming and batch processing for immediate data consistency.
- Distributed Databases - Manages a distributed database cluster ensuring horizontal scalability and fault tolerance.
- Parallel Query Execution - Distributes complex queries across multiple data nodes and merges partial results via a central broker.
- Online Analytical Processing - Executes high-performance slice-and-dice queries on massive volumes of data for rapid trend and pattern identification.
- Real-Time Data Streaming - Ingests continuous data streams into the database for immediate visibility into real-time business metrics.
- Stream Processing Engines - Ships a streaming ingestion engine for the continuous analysis of arriving high-velocity data.
- Inverted Index Engines - Employs inverted indexes and bitmaps to enable rapid filtering and set operations on large datasets.
- Data Partitioning - Divides datasets into time-based chunks to enable parallel processing and querying across the cluster.
- Distributed SQL Querying - Supports standard SQL for querying, monitoring, and managing distributed database clusters.
- SQL Query Interfaces - Provides standard SQL interfaces for querying datasets and analyzing distributed data.
- Distributed Coordination Systems - Utilizes an external consensus store to manage cluster metadata and coordinate node health.
- Cluster Management - Provides centralized management and monitoring of cluster health, segments, and services via SQL system tables.
- Big Data Storage - Column-oriented distributed store for interactive analytics applications.
- Distributed Databases - Fast column-oriented distributed data store for real-time analytics.
- Time Series Databases - Distributed data store for powering interactive analytical applications.
- Real-time Monitoring - Distributed, column-oriented data store for real-time analytics.
