Streaming SSR
Each section renders independently with
<Suspense>. El HTML is streamed progressively to the browser without waiting for everything to finish.Loading DB data...
Loading external API...
Loading ML inference...
What does this show?
- ‣
<Suspense>wraps each async section independently - ‣The HTML shell is sent to the browser immediately
- ‣Each section appears when its data is ready (200ms, 1s, 3s)
- ‣Skeletons are shown as fallback while each promise resolves
- ‣No waterfall: all 3 data sources load in parallel