Skip to main content

Global Lineage View

The Global Lineage page provides a comprehensive visualization of all data products and their relationships within a space.

Accessing Global Lineage​

Navigate to Lineage in the main sidebar to open the space-wide view.

URL: /lineage

Graph Overview​

The global view displays every data product in the space as a node, with all lineage relationships shown as connecting edges. The graph automatically arranges these elements using a hierarchical layout that flows from sources to consumers.

Layout Algorithm​

The graph uses topological sorting to determine positioning. It identifies root products (those with no upstream sources) and places them at the far left (Level 0). Subsequent products are assigned to levels based on their dependency depth, with nodes at each level centered vertically. Edges are routed using a step-style for a clean visual flow.

Visual Hierarchy​

The layout provides a clear visual hierarchy: source data sits at the far left, transformation layers occupy the middle, and consumer endpoints like dashboards and exports are positioned at the far right.

Searching and Locating​

To find specific products in a large graph, use the search bar in the graph header. Typing a product name filters results in real-time; selecting a match centers the viewport on that product.

Keyboard Navigation​

You can cycle through multiple search matches using the Enter key, with a counter showing your position in the results (e.g., "1 of 5").

Viewport Centering​

When you locate a product, the graph automatically centers on it and adjusts the zoom level for optimal visibility. The selected node is highlighted to make it stand out visually against the rest of the graph.

dbt-Style Selectors​

For advanced navigation, the graph supports dbt-style selector syntax in the selector input field.

Syntax Patterns​

You can select a single product by name (product_name), or include its context: +product_name selects the product and all its upstream sources, while product_name+ includes all downstream consumers. To see the full lineage, use +product_name+. You can also limit depth, such as +2,product_name for two levels of upstream dependencies, or product_name,+1 for one level of downstream.

Using Selectors​

Enter a selector to filter the graph; the view updates to show only matching nodes and their relevant connections. Press the Clear (X) button to return to the full graph view.

Examples​

Entering +customer_orders+ shows customer_orders along with all its sources and consumers. Using +2,daily_sales focuses on daily_sales and its upstream dependencies up to two levels deep.

dbt Users

If you're familiar with dbt's --select syntax, you'll feel right at home with these selectors.

Interactive Highlighting​

Path Focus​

Hovering over any node or edge highlights the connected path. In-path elements remain fully visible, while the rest of the graph dims to 25% opacity in grayscale. This immediate visual focus makes it easy to trace specific pipelines even within complex graphs.

Node Interaction​

Clicking Nodes​

Clicking any node triggers a fast, single-page navigation to that product's detail page. You can use your browser's back button to return to the lineage view preserving your state.

Node Information​

Nodes display the product name and a color-coded type icon, providing a quick visual reference for the product's role in the pipeline.

Orphan Products​

Orphan Cluster​

Products with no lineage connections are grouped in a separate "Orphan Cluster" positioned below the main graph. This area uses a grid-based layout to clearly separate these disconnected items from the main pipeline.

Why Orphans Exist​

A product might appear here if its lineage hasn't been curated yet, if it's a standalone data source, or if it was newly created. To integrate an orphan, open its detail page and add connections via the Lineage tab.

Graph Controls​

Controls in the graph corner allow you to Zoom In and Zoom Out to adjust the view, Fit View to automatically frame all nodes, and Reset to return to the default zoom and position.

Minimap​

A minimap in the corner provides an overview of the entire graph, showing your current viewport position. You can click or drag within the minimap to quickly pan across large graphs.

Performance Considerations​

Large Spaces​

For spaces with many products, the initial load may take a moment. Using search and selectors to focus on relevant areas, or filtering by tags and types, can help manage complexity.

Browser Performance​

If the graph feels slow, try reducing the number of visible nodes using selectors, closing other processor-intensive browser tabs, or refreshing the page to reset the graph state.

Use Cases​

Pipeline Architecture Review​

The global view is ideal for reviewing overall data architecture. It helps identify data silos (unconnected clusters), locate bottleneck products with excessive downstream consumers, and spot missing connections in your pipeline.

Onboarding​

New team members can use the global view to understand major data flows, identify key source and consumer products, and use selectors to explore specific pipelines without getting lost.

Impact Assessment​

Before making changes, find the product in the global view and use +product_name+ to visualize its full impact. This helps identify all affected downstream consumers so you can plan communication and coordination effectively.

Best Practices​

Regular Review​

Periodically review the global lineage to ensure connections are up to date, identify orphan products that require curation, and spot any unexpected patterns or gaps in your data architecture.

Collaborative Curation​

Lineage is most valuable when complete. Encourage teams to maintain their products' lineage, include lineage review in onboarding processes, and use the global view in governance meetings to drive visibility and accountability.