Key Concepts

Before we get started with deploying on TrueFoundry, it’s important to understand a few key concepts - Cluster, Workspace, Repository, Environment.

​

Cluster

A Cluster in TrueFoundry refers to an actual Kubernetes cluster in one of your cloud accounts or on-prem.

​

Workspace

A TrueFoundry workspace refers to namespace in Kubernetes. A cluster can have multiple workspaces. Each workspace is a logical separation within a cluster so that different teams, applications, and environments can sit within a cluster.

For e.g, let’s say there are three teams within a company. Team1 manages application1, application2, team2 manages application3, and team3 manages application4 and application5. Each of the applications further has three environments - dev, staging, and production. One way to organize the workspaces can be:

We can also have the dev and prod workspaces in one cluster if you want to structure them like that.

​

Repositories (also called MLRepo)

Repositories in TrueFoundry are repositories to store models, data, artifacts and prompts. They are backed by a blob storage like S3, GCS, or Minio. Like in Git repositories, you can configure access to read, or write to different users, you can similarily configure access control in TrueFoundry repositories.

You can also grant access of repositories to certain workspaces. Once a workspace has access to an ML repo, all applications inside the workspace can then read or write to the MLRepo depending on whether the workspace has viewer or editor access. This way, we don’t need to inject any keys and this makes the entire system much more secure.

In the case above, workspace1 has access to MLRepo1 and hence team1 can access the assets in MLrepo1. This way, you can also divide datasets and model access across teams and workspaces.

​

Environment

An environment is a tag applied to workspaces to categorize them based on factors like development stage - for example: dev, staging, production. We create three environments initially dev, staging and prod - the names of which you can change at any point of time. A cluster can be associated to multiple environments but a workspace can only be associated to one environment. For each environment, you will need to define its for production and if it needs to be optimized for cost or availability. These settings are used to power the recommendations to improve your applications.

​

FQN

FQN stands for Fully Qualified Name and is a more human-readable version of the id in the TrueFoundry platform. Every entity - be it a cluster, workspace, respository, secret, model has a FQN. You can copy the FQN from the UI for every entity.