High Performance Data Visualization for the Web
summary
Builds a high performance web frontend using Perspective and demonstrates realtime filtering sorting pivoting and aggregation on large and streaming datasets, with a WebAssembly bound C++ engine and bindings for Python and JavaScript plus Rust plugins. Uses Apache Arrow for fast diffs and supplies ready websocket integrations like the FastAPI websocket handler so servers push consistent pivots to clients. Live demo shows subsecond updates (0.1s) with real time aggregations, works in-browser until browser memory limits and runs in production at several organizations.
AI-written summary · about
chapters (8)
- 0:00Introduction
- 1:26Web frameworks pain points
- 8:49Introducing Perspective
- 10:53Technical architecture
- 12:38Live demo
- 17:40Real-time streaming
- 20:30User exploration
- 21:13Q&A
related talks
29mHigh Performance Data Visualization for the Web
24mCreating Interactive Data Visualizations for the Web
30mSeeing the needle AND the haystack: single-datapoint selection for billion-poin...
47mUsing Julia & D3 to analyse web performance data
40mThe Power & Performance of Phoenix LiveView
03mVisualizing data with JavaScript at Twitter
29mParallel computing in ReactJS: Provide a pleasant user experience
31mStreaming React for Faster UIs