Embedded devices such as RaspberryPI don't have enough power for running hadoop jobs. It is a really complex peace of software. Instead of trying to adapt it I decided to use more "lightweight" solution which is OpenMPI map-reduce. MPI was designed for distributed computations, so why not run map-reduce framework on it? Let's do it!
This 3rd article is going to cover post-installation process, where we'll install golang + docker +fleet + etcd. As most of these tools is made for 64bit systems we will have to modify them a bit. The set of tools is used in coreos for managing containers in a cluster, so that is a perfect tool for us!
Most of Raspberry like ARM devices have system image written to sdcard. Whenever we need to update base image we need to take out memory stick, write image and put it back. Now imagine that you have 8 boards and you want to update all of them (e.g your current system is crashing). So, you won't do it manually, will you?
Have you ever dreamed of building your own cluster? Running your own service on your own machines? Now with ARM boards it is possible! You don't need much space or excessive amount of money. In this article you may see how I did it and how to control power of each ARM device separately.