This post is a summary of the plan for the Raspberry Pi Cluster project.
Progress so far
So far I have been creating a variety of scripts to see what you I can easily do with the cluster project. At the moment these scripts are spread across the nodes and not easily accessible.
Current examples running on the cluster
- All nodes connected via sockets
- Reporting node statistics to master (cpu, ram, swap)
- Shared distributed filesystem
- Multi node file download and organisation
- Distributed md5 hash generation
- Web UI management interface
These have all been written as proof of concept scripts to ensure they work on the cluster. Now I have found a number of projects the cluster can easily perform I am writing them up.
The repository and plans
Using a GitHub repository I will be describing steps to build up the cluster. This will go from a single node logging that it is alive to many nodes running many processes. Using the various examples I have written already I will describe how to build it using code examples.
The Github Project for this is https://github.com/chewett/RaspberryPiCluster
Each blog post will link to the GitHub project with full code available. The blog posts will explain what we are doing and what it will be used for. Each subproject will build in the previous to build up a full codebase for our cluster.
Making all of the information opensource will hopefully let others play with clustered computing on a small scale.