To provide fluid, visually rich user experiences on infinite digital canvases, Atlassian developed a custom GPU-powered rendering pipeline in Confluence whiteboards. This approach balances high performance and graphical quality by converting shapes into triangles and applying mathematical transformations to achieve dynamic rounded corners during real-time interactions.

  • GPU-friendly triangulation converts shapes into triangles.
  • Mathematical transformations create and maintain rounded corners dynamically.
  • Custom WebGL pipeline sustains smooth zoom and pan at 60+ FPS.

What happened

Atlassian has introduced a new rendering technique in Confluence whiteboards that uses a GPU-accelerated graphics pipeline built on WebGL to handle thousands of shapes and connectors efficiently. Rather than relying on standard HTML and CSS, the company triangulates shapes into GPU-friendly triangles and applies complex mathematical transformations to create smooth rounded corners on various shapes. This ensures consistent frame rates above 60 FPS, even on large, detail-rich boards.

This approach addresses the challenge of rendering many scalable shapes with visually appealing rounded edges, such as polygons and circles, by approximating complex curves with collections of triangles. To achieve this, Atlassian uses linear transformations to stretch and rotate quarter circles to fit the corners of shapes, preserving the quality and responsiveness of the interface during user interactions like panning and zooming.

Why it matters

Confluence whiteboards are designed for real-time collaboration, requiring seamless user experiences regardless of the board's size or complexity. The traditional web rendering stack can struggle with the performance demands when displaying thousands of interactive shapes simultaneously with dynamic visual effects like rounded corners. Atlassian's GPU-powered solution directly tackles these limitations to maintain fluidity and responsiveness.

By leveraging the GPU’s native strength in processing triangles and applying geometric transformations on the client side, Atlassian has set a new performance standard for web-based diagramming and brainstorming tools. This ensures users can focus on teamwork and creativity without frustration from lag or diminished visual fidelity, ultimately improving adoption and satisfaction.

What to watch next

As Atlassian refines its GPU rendering pipeline, future developments may include more sophisticated shader effects or optimizations to further reduce CPU and GPU workload. Observers should watch for the integration of more advanced shape types or real-time effects that maintain performance while expanding creative possibilities for users.

Additionally, this technology may influence wider adoption of WebGL-based rendering in SaaS collaboration products, where balancing performance and visuals is critical. Competitors and other productivity platforms may explore similar GPU-centric approaches to keep pace with user expectations for smooth, dynamic interfaces.

Source assisted: This briefing began from a discovered source item from Atlassian Blog. Open the original source.
How SignalDesk reports: feeds and outside sources are used for discovery. Public briefings are edited to add context, buyer relevance and attribution before they are published. Read the standards

Related briefings