FastPubSub

Ultra-low latency distributed messaging network

News & Blog

Project updates, technical insights, and development progress

dashboard.fastpubsub.com is live

Today we launched dashboard.fastpubsub.com — a real-time monitoring dashboard for the FastPubSub overlay network.

The dashboard shows a live world map of all active PoPs (Points of Presence) and the connections between them. You can see in real time how our overlay routes compare to direct internet paths: which routes are faster through the overlay, where the public internet has problems (unreachable hosts, abnormally slow links), and how much latency we save on each hop.

  • Live latency map — every node and every measured route between them, updated in real time via WebSocket.
  • Overlay vs Direct comparison — each route shows the direct internet RTT and the overlay RTT side by side, with the saving in milliseconds and percent.
  • Network anomaly detection — unreachable routes (ping timeout) and abnormally slow routes are highlighted separately so you can see where the public internet is struggling.
  • Adjustable filters — sliders to set the minimum saving threshold and the slow-route threshold, so you can focus on the routes that matter.

The dashboard itself is built on top of FastPubSub — it uses our own WebSocket pub/sub infrastructure to stream measurement data from every PoP to the browser. It is both a monitoring tool and a live demo of the network in action.

We are also getting ready to build the first SDKs. The goal is to provide simple, lightweight client libraries for major languages so developers can integrate FastPubSub into their applications with just a few lines of code. More details on SDK plans coming soon.

FastPubSub MVP is taking shape

We are excited to share that our MVP is actively evolving and already has core functionality working end-to-end. Here is what we have operational today:

  • Overlay Network — Our custom overlay mesh built on QUIC/UDP is up and running. Brokers discover each other, establish encrypted tunnels, and form a resilient topology independent of public internet routing.
  • Latency-Based Routing — Brokers continuously measure round-trip times between each other and compute the fastest paths using a Distance Vector algorithm. Traffic is routed through the lowest-latency path in real time.
  • Access Tickets — Our authentication layer is in place. Clients receive access tickets that grant them permission to publish or subscribe to specific channels. This ensures that only authorized participants can interact with the network.

The foundation is solid and we are now focused on hardening these components, improving observability, and preparing for the first external testers. Stay tuned for more updates as we move toward a public beta.