Performance Engineering in Dynamic Web
summary
Platform serves pages with over half a million pixels and almost 3,000 widgets on a single HTML document, so performance relies on concrete choices: multi‑level caching (CDN + Redis + server‑side rendered HTML), chunked JS with HTTP/3 and viewport‑first loading, lazy hydration driven by a custom IntersectionObserver, LQIP plus on‑demand image variants with srcset/sizes, fontface API with variable fonts, animations moved to the compositor via the Web Animations API, and a signals‑based state system for near‑zero update overhead.
AI-written summary · about
chapters (14)
- 0:00Introduction to talk
- 1:01Evolution from static to dynamic web
- 3:28Web 2.0 and new challenges
- 4:52Mobile era and responsive design
- 6:40Defining dynamic web today
- 8:06Examples of complex dynamic projects
- 9:09Multi-level caching strategies
- 10:33Optimizing script loading
- 11:16Partial rendering and lazy hydration
- 11:58Image optimization techniques
- 13:02Custom font performance
- 13:22Web Animations API for smooth effects
- 14:27State management with Signals
- 15:50Future of performance engineering
related talks
46mWeb Performance in 2018
33mRendering on the Web: Performance Implications of Application Architecture
25mBetting on the web: Ensuring fine-tuned performance for dynamic web apps
21mEvolution of Web Performance
16mThe Evolution of Web Performance
27mAngular Performance and Core Web Vitals in 2024
15mPersonalization, Performance, and Probably Dynamic Content
54mPushing Pixels