Browser memory is limited. Scramjet’s browser build automatically overflows to IndexedDB when backpressure builds up, allowing terabyte-scale streaming on a laptop.
Unlike a standard browser, Scramjet is built on a headless host—Node.js. It runs on the V8 JavaScript engine, but it bypasses the heavy DOM rendering layer.
Here is the high-level architecture of how Scramjet works: scramjet browser work
We can use reduce to accumulate data into an HTML string and update the UI incrementally.
const outputDiv = document.getElementById("results");
DataStream.from(response.body) .JSONParse() .map(item =>Browser memory is limited<div class="card">$item.name</div>) .reduce((html, card) => html + card, "") .then(html => outputDiv.innerHTML = html; );
| Operation | Node.js native streams (ms) | Scramjet (ms) | Improvement | |-----------|----------------------------|---------------|--------------| | 1M lines filter + map | 1450 | 920 | 36% faster | | 100k objects groupBy | 890 | 540 | 39% faster | | Backpressure test (10x burst) | dropped 12% | 0% drop | Stable | | Operation | Node
Test environment: Node.js 18, 4 CPU cores, 8GB RAM.
A real scramjet (Supersonic Combustion Ramjet) doesn’t carry heavy oxygen tanks. It flies so fast that it rams air into its engine at Mach 5+. Combustion happens instantly in the supersonic airflow.
A Scramjet Browser operates on the same principle: continuous, supersonic data flow. It eliminates the "stop-and-start" nature of navigation.