Welcome to Cloudster

Cloudster is a distributed k-means algorithm implementation. It aims at partitionning huge quantities of data into clusters using user-provided metrics.

Key features

• Written with .NET and Windows Azure platforms
• Generic implementation
• Heavily scalable implementation
• BSD licensed

Getting started

Download the latest version here and go to getting started to learn how to install it, run samples and use it for your own project.

Use cases

• DNA sequences comparison
• Text analysis
• Image comparison

Resources

This project is hosted on SourceForge : Project page.
Other k-means implementations : SourceForge list.