Scalable Data Visualization
summary
Uses React.js to render real, shipped visualization components over a streaming pipeline that parses a 21 GB Bitcoin blockchain into 45 million transactions, aggregates with sketch streaming aggregation and binning into ~47 KB updates every two seconds, and sends them over websocket/XMLHttpRequest to a browser that calls setState and renders SVG or canvas. Architecture uses a stateful server with Flux-like dispatcher, pages raw rows around browser scrolling limits, and reports working up to a terabyte of data.
AI-written summary · about
chapters (19)
- 0:00Start
- 2:26Building a predictive application
- 3:15Why Data Visualization?
- 3:45Real world example: Bitcoin transactions
- 10:41Visualizing large data: challenges
- 11:49Data Visualization Pipeline
- 13:02Compute close to the data Server
- 13:32"Flux" on the server
- 14:43Scalable aggregation
- 17:20Example (server)
- 17:44Example (client)
- 19:20Visual elements, cont'd.
- 19:53Anatomy of d3.js
- 21:27Example (parent component)
- 21:48Example (child component)
- 22:05Choosing a render target
- 23:00Considerations
- 23:49Render targets compared
- 25:06The Dato visualization architecture
related talks
46mExplorations in Interactive Visual Analytics
15mThe Future of Data Visualization on the Web
43mThe Future of Data Visualization on the Web
21mData Visualizations with React + D3
21mVisualizing the Ethereum Blockchain
41mVisualizing big data
03mVisualizing data with JavaScript at Twitter
42mData visualizations on the web