Back to Gap Map Infrastructure / Storage & Retrieval

TileDB

TileDB
open source / Overall score: 2.4

TileDB, also called TileDB Embedded, is an embeddable C++ storage engine for dense and sparse multi-dimensional arrays, which it also uses to store dataframes. Data is stored in compressed tiles on local disk or on S3, Google Cloud Storage and Azure, with parallel I/O, versioning and time travel, and it is reached through C, C++, Python, R, Java, Go and C# APIs and integrations with Spark, Dask and GDAL. Its maker, TileDB, Inc., trades as Tile.ai and sells a separate enterprise data platform under that name.

Openness

5 high confidence
5.0
license
MIT (OSI)
source
public
core features withheld
no — the Tile.ai enterprise platform is a separate product, not part of this repository

The storage engine is MIT-licensed and builds in full from its public repository. TileDB, Inc. sells a separate enterprise data platform, now branded Tile.ai, which it describes as built by the team behind the open-source engine; nothing in the engine is held back for it.

Adoption

2 medium confidence
2.0

PyPI downloads of tiledb, the Python API, whose wheels carry the engine itself. Installs of the C++ library from conda-forge and use through the R, Java and other APIs are not counted.

Capability

3 medium confidence
3.0

TileDB is a complete embeddable store for arrays and dataframes, with versioning, encryption and direct reads and writes against cloud object storage, the same kind of single store as Deep Lake. It runs inside the application that calls it and has no clustered serving or built-in search; vector search lives in a separate TileDB project.

  • https://raw.githubusercontent.com/TileDB-Inc/TileDB/main/README.md recorded 2026-11-12

    README features: "dense and sparse arrays", "dataframes and key-value stores (via sparse arrays)", "Cloud storage (AWS S3, Google Cloud Storage, Azure Blob Storage)", "Multiple compression, encryption and checksum filters", "Parallel IO", "Data versioning (rapid updates, time traveling)", integrations "(Spark, Dask, MariaDB, GDAL, etc.)"; "It is an embeddable C++ library"; TileDB-Vector-Search listed as a separate application-specific package.

Verified 2026-11-12