heavyai/heavydb

Health v2 · maintenance only

Adoption not part of the score

3057 stars · 457 forks observed · 2026-08-28

What it is AI-extracted, prompt v1, taxonomy v1, 2026-08-30, confidence not recorded

HeavyDB (formerly MapD/OmniSciDB) is an open-source SQL-based, relational, columnar database engine that uses CPUs and NVIDIA GPUs to query multi-billion row datasets in milliseconds. It features multi-tiered caching across storage, CPU memory, and GPU memory, plus JIT query compilation, without requiring indexes, pre-aggregation, or downsampling.

Use cases

• query billions of rows in milliseconds with SQL
• run interactive analytics on GPU-accelerated hardware
• visualize large datasets in real time without downsampling
• run OLAP workloads on hybrid CPU/GPU systems
• avoid indexing and pre-aggregation for ad-hoc analytical queries
• accelerate machine learning feature queries on large tables

When to choose

• you need sub-second interactive queries over multi-billion-row datasets
• you have NVIDIA GPUs and want to exploit them for analytics
• you want a columnar SQL engine without index maintenance overhead
• you need real-time dashboards and visualization backends

When to avoid

• you need a general-purpose transactional (OLTP) database
• your deployment is Windows or macOS only (prebuilt binaries target Linux)
• you lack GPU hardware and need ARM support beyond experimental
• you need a fully managed cloud database with vendor support

Facets

service · maturity active

database search-engine data-visualization machine-learning databases analytics big-data data-visualization gpu-computing cloud olap columnar gpu-accelerated jit-compilation real-time-analytics sql-database mapd omnisci sql linux docker gpu

Member repositories

For agents

markdown · JSON · MCP: product_card(name="heavyai/heavydb")