apache/doris

Apache Doris is a real-time analytics and hybrid search database for AI agents.

Apache Doris – Real‑time Analytics & Hybrid Search Engine

What it is

  • An open‑source, massively‑parallel processing (MPP) database designed for sub‑second, high‑concurrency SQL queries.
  • Handles structured tables, full‑text, JSON, and vector data in a single engine, making it useful for AI‑related workloads such as similarity search and retrieval‑augmented generation (RAG).

Key capabilities

Capability Why it matters
Real‑time analytics Streams data in, transforms it incrementally, and serves queries in < 1 s even under heavy load.
Lakehouse analytics Executes fast SQL over open table formats (Iceberg, Delta Lake, Hudi) stored in object storage, letting you query data where it lives.
Hybrid search Native SQL support for full‑text, JSON, and vector similarity search, enabling AI‑driven retrieval and recommendation use‑cases.

Typical use cases

  • Customer‑facing analytics – interactive dashboards that must respond instantly for external users.
  • Data warehousing – a single, real‑time warehouse that spans multiple business domains.
  • Observability – ingesting and querying massive streams of logs, metrics, and events with SQL.
  • AI / RAG – storing embeddings alongside traditional data and performing vector similarity joins directly in SQL.

Architecture snapshot

  • Supports coupled (compute + storage together) and decoupled (stateless compute nodes reading from shared object storage) deployments.
  • Decoupled mode lets you scale compute independently and isolate workloads, a pattern common in modern cloud data platforms.

Ecosystem & integrations

  • Connectors for Flink, Spark, Kafka, and a Kubernetes Operator for containerised deployments.
  • Fits into the broader data stack, linking upstream ingestion systems to downstream BI, AI, and observability tools.

Getting started

  • Quick‑start guides, binary downloads, and Docker‑based source compilation are provided on the official site.
  • Community support via Slack, GitHub Discussions, and a weekly community report.

Why it matters for AI practitioners

  • Stores and queries vector embeddings alongside relational data without a separate vector database.
  • Enables low‑latency retrieval for RAG pipelines, recommendation engines, and semantic search directly from the same SQL layer used for analytics.

License

  • Apache License 2.0 (with a note that some third‑party components have differing licenses).

All details are taken from the project's README; no additional features are inferred.

Related

  • Project
    manticoresoftware/manticoresearch

    A high-performance, open-source search database that provides full-text, vector, and hybrid search capabilities as a fast alternative to Elasticsearch.

  • Project
    infiniflow/infinity

    An AI-native database providing high-speed hybrid search across dense and sparse embeddings, tensors, and full-text for LLM and RAG applications.

  • Project
    elastic/elasticsearch

    A distributed search and analytics engine and vector database that enables near real-time search and analysis over massive datasets for AI and monitoring applications.

  • Project
    milvus-io/milvus

    Milvus is a high-performance, distributed vector database built for scaling AI applications that require efficient search and organization of massive amounts of unstructured data.

  • Project
    redis/redis

    Redis is a high-performance, in-memory data structure server and vector query engine used for caching, real-time analytics, and GenAI memory.