Scales from laptops to clusters
Dask is convenient on a laptop. It installs trivially with conda or pip and extends the size of convenient datasets from “fits in memory” to “fits on disk”.
Dask can scale to a cluster of 100s of machines. It is resilient, elastic, data local, and low latency. For more information, see the documentation about the distributed scheduler.
This ease of transition between single-machine to moderate cluster enables users to both start simple and grow when necessary.