A small atlas of the moving parts behind distributed data.
The systems in CS 544 are easy to draw on a chalkboard and hard to hold in your head all at once. This is a quiet attempt at the opposite — diagrams you can poke at, with a few sliders and a lot of restraint.
§ 1The premise
Most explanations of Cassandra rings, Kafka partitions, and HDFS writes are static — a single diagram, frozen in one configuration. But the questions a student actually has are dynamic: what happens when I change the replication factor?; what happens when two messages share a key?; where do the bytes flow if a DataNode goes down?
Each page below is a single, focused interactive figure that answers one of those questions. Drag a key around the Cassandra ring and watch the replicas update. Send a message into a Kafka topic and watch which partition it lands on. Push a file into HDFS and watch the bytes flow.
No telemetry. No build step. Just SVG and a little JavaScript.
§ 2The figures
§ 3How to read the figures
Each page is structured the same way. A short paragraph in plain English. The figure itself, with the controls underneath it. A readout strip that updates as you change things. A caption that points out what to notice. And, beneath it all, a short “the rules” section that distills the figure into the one or two facts that are most likely to be tested.
The figures are not simulations of real clusters. They are faithful to the algorithms — Cassandra’s ring walk really does go clockwise; Kafka really does hash key % n; HDFS really does pipeline — but the numbers and timings are schematic.
§ 4Colophon
Built by Nils Matteson, a CS 544 student in spring 2026, mostly as a way to study for the final by building the diagrams I wished I had been studying from. With thanks to Tyler Caraza-Harter, whose lectures and exam questions are the source material for nearly every interaction here.
Set in EB Garamond and JetBrains Mono. Source and issues at github.com/matteso1/CS544Visualizations .