SPACE Architecture: MicroVMs and Performance Metrics
perplexity_ai · x · 2026-07-16
Perplexity shared more details on the underlying architecture and performance of the SPACE sandbox platform:
- Architecture: SPACE decouples user sessions from the underlying sandbox. Each task gets a disposable Firecracker microVM that is destroyed upon completion. Rolling snapshots preserve live memory and files, allowing sessions to be paused, resumed, or branched across sandboxes at any time.
- Performance: Under identical production traffic, SPACE reduced median sandbox creation latency from 185 ms to 60 ms, and P90 latency from 447 ms to 89 ms. Last week, the system handled millions of sandbox creations and tens of millions of reconnections.
Related event: Perplexity Unveils SPACE, a Sandbox Built for Long-Running Agents(8 posts)→
More from coding & agent
- Building a Secure AI Agent Gateway: Self-Hosting OAuth for Multiple SaaS Apps — Defiant_Cod_2654 · 2026-07-22
- Rowboat launches as an open-source, local-first AI coworker with memory — ycombinator · 2026-07-22
- Scoble says AI “loops” really means long-running multi-agent workspaces — Scobleizer · 2026-07-22
- Kimi Code opens a waitlist as Moonshot rolls out its coding product — Fabulous_Bonus_8981 · 2026-07-22
- Open-source runtime lets each repo define its own AI code reviewer — ibabufrik · 2026-07-22
- Indie Dev Asks: What's Actually Broken in Your AI Agent's Memory Today? — AcceptableTime7937 · 2026-07-22