Dask
Creating and spinning up Dask clusters
If you are just taking a look at Saturn, you can easily sign up for a free trial of our hosted product. This way you can dive right in without needing to do any setup or installation.
After logging in, you’ll see the Project page. From this page you can open one of the example projects or create a new project. Let’s look at an example. Click on “examples-cpu” to go to the project details page.
This page shows all parts of the project and allows you to start and stop all the resources associated with the project.
To run through the example notebooks we first need to start up the Jupyter server.
Start the Jupyter server by clicking the play button next to “Jupyter Server”. Your Jupyter should be ready in 1-10 minutes. Once it is, click “Jupyter Lab” to read through and run the example notebooks.
All the details on how to execute these notebooks is included in README.md files.
The examples use dask-saturn
, a library that comes pre-installed in all Saturn environments. You’ll see that they spin up Dask clusters by executing something like:
from dask_saturn import SaturnCluster
cluster = SaturnCluster()
cluster
As you run through the examples you can inspect the Dask diagnostic dashboard to get a sense of how tasks are distributed between workers.
Creating and spinning up Dask clusters
Sign up for Saturn in the AWS marketplace, and deploy Saturn into your AWS account.
Inviting your team to Saturn