I'm working on a Hadoop / Native (Distributed File System vs Native File System) comparison / benchmarking project. There are a few conjugates of scripts to be made such as creating copy, paste, duplicate, move, delete mass files on both file systems; simulate a video broadcast environment (both client and server side, not a detailed thing). Those should be able to run with more than one instances. In the end, i'll run PowerApi to measure the power consumption on the computer. It looks easy but I expect you to help me with integration.
* The aim is compare the energy consumption for each jobs on different file systems. (Hadoop DFS vs Native)
OS will be Ubuntu, version is probably the latest.