Gigantum Documentation

You'll find comprehensive guides and documentation to help you start working with the Gigantum platform as quickly as possible. If you get stuck, then there is help for that too. Just reach out to the team or get help from the community using our our Spectrum channel.

Get Started    Changelog

Gigantum Hub

A conceptual overview of the Hub.

Gigantum Hub is a service for storage, sharing and collaboration that provides public and private spaces for storing, posting and finding Projects and Datasets. It also provides cloud-based compute for the purposes of exploration and inspection of posted Projects without having to install the Client.

What does the Hub do?

Gigantum Hub provides three things.

  • Storage for backup and transfer of work done in the Gigantum Client so users can compute across different machines and still have full access to versioned work.

  • Public and private sharing for dissemination and collaboration so users can share work with the world, find new things and collaborate.

  • Limited cloud compute resources to test, inspect and explore Projects.

A Project in Gigantum Hub

Storage and backup

Gigantum Hub is analogous to GitHub, but tailored for data science projects and the diversity of interests and skills of data scientists. Users can sync work between local repositories and the Hub. All data is encrypted, backed up, and available to download anywhere.

The Hub interface makes it easy to inspect a Project's data, code, environment configuration and user attributed history without installing anything. For public Projects and Datasets, you even need an account to view content. For an example of this, check out a public Project on Gigantum Hub here.

Public and private sharing

When users publish local work to Gigantum Hub, they choose to make it public or it private. Public Projects and Datasets are visible on the Explore page and anyone can download or copy them. Private ones can only be viewed by collaborators with permissions. Collaborators can be added to either public or private Projects with a variety of privileges.

If you want to see the Projects and Datasets for a user, you can just go to their namespace. This will show public Projects and Datasets along with any private Project or Dataset to which you've been added as a collaborator. For example, our CTO's namespace is here.

Computing and using the Client in Gigantum Hub

While the Client runs locally, we also provide computational resources in our cloud because:

  • Inspecting data science doesn't just reduce to looking at files. It means computation.
  • Users may need to access and use cloud computing while working on machines that don't have the Client installed locally.

Users can run code for any Project that they have access to on the Hub by going to the files tab and clicking the Launch button in the right hand corner. A Jupyter instance (Classic or Lab) will load in a new tab, but the environment may take some time to build depending on the configuration and if it has been built before in Gigantum Hub.

Each free Gigantum account provides a limited amount of compute time per month that can be upgraded. Learn more about [Gigantum Hub Resource Limits here](doc:gigantumhub-resource-limits.

An example project Launching in Gigantum Hub


Caveats on computing in Gigantum Hub

Computing in the Hub has some caveats, the main one being that work done there is ephemeral. This means that when you close a session (times out after 20 minutes of not running a Project) you can lose your work unless you've properly synced it with Gigantum Hub. So, if you are doing something you care about in the Hub, then you should make sure to open the Client and sync.

Updated 28 days ago

Gigantum Hub

A conceptual overview of the Hub.

Suggested Edits are limited on API Reference Pages

You can only suggest edits to Markdown body content, but not to the API spec.