Console tour
A guided tour of the DataHub web console, the home dashboard and every section in the top navigation.
Guided tutorial
The console teaches itself. Where to start the built-in guided tutorial, the six chapters it walks you through, the four industry scenarios it can speak in, what it creates in your tenant and how to remove it afterwards.
Data sets
What a data set is in DataHub, how to design good data set boundaries, and why they double as the access-control boundary.
The resource graph
Creating resources, connecting them with relationships, labelling them, and using the network view to trace dependencies.
Time series
Creating and managing time series in DataHub, units, value types, step series, and connecting a measurement to the thing it measures.
Insights, charting and live data
Plotting time series in the DataHub console, comparing several at once, choosing time ranges, and streaming live readings.
Relationship analysis
Discovering which time series move together, which one leads, and whether the relationship is dependable, explained in plain language.
Events
What events are in DataHub, how the event log works, and why anchoring events to resources is what makes an investigation fast.
Data subscriptions
Subscribing to data so it is pushed to you the moment it arrives, millisecond-scale delivery, the use cases that only exist once latency stops being a schedule, and a worked sliding-window example that turns four hundred nuisance alarms into one event.
Files
Attaching documents and images to your model, P&IDs, datasheets, inspection photos and certificates, with folders, data sets and a recoverable trash.
Change data capture
Connecting a source database directly to DataHub so its changes flow in automatically, what CDC is good for, and when to use something else.
Functions
Modelling computations in DataHub today, and the execution engine on the roadmap, data cleaning, transformation, feature extraction and streaming computation.
Policies
Naming policies, enforced on every write, plus the lifecycle, access and requirement rules recorded today and the enforcement that is on the roadmap.
Building applications on DataHub
Using the platform as a foundation for your own applications, what an app inherits for free, the first application worth building, and the design rules that make apps compound the model instead of siloing it.
Using MCP
What the Model Context Protocol is, why it changed what AI assistants can do, and how to point one at your own DataHub so it answers from your operation instead of from general knowledge. The two servers, the 38 tools, and how access is decided.
Building AI agents
What an AI agent actually is, what you build against DataHub versus what you get for free, the four limits worth setting, and seven worked examples of what agents can already do today, across oil and gas, wind, shipping, data centres, defence, markets and healthcare.