Sparse and Dense Multidimensional Array Storage Engine for Data Science

The data science storage engine 'TileDB' introduces a powerful on-disk format for multi-dimensional arrays. It supports dense and sparse arrays, dataframes and key-values stores, cloud storage ('S3', 'GCS', 'Azure'), chunked arrays, multiple compression, encryption and checksum filters, uses a fully multi-threaded implementation, supports parallel I/O, data versioning ('time travel'), metadata and groups. It is implemented as an embeddable cross-platform C++ library with APIs from several languages, and integrations.


Reference manual

It appears you don't have a PDF plugin for this browser. You can click here to download the reference manual.


0.8.2 by Dirk Eddelbuettel, 3 months ago

Report a bug at

Browse source code at

Authors: TileDB , Inc.

Documentation:   PDF Manual  

MIT + file LICENSE license

Imports methods, Rcpp, nanotime

Suggests tinytest, rmarkdown, knitr, BiocStyle, curl, bit64

Linking to Rcpp

System requirements: cmake (only when TileDB source build selected), git (only when TileDB source build selected); on x86_64 platforms pre-built TileDB Embedded libraries are available at GitHub and are used if no TileDB installation is detected, and no other option to build or download was specified by by the user.

See at CRAN