ANALYZING A SPECIALTY COMPUTER LAB USING RASPBERRY PI

Thomas A Robbins (Professor Joel Hollingsworth) Department of Computing Sciences

The essential question addressed by this research is: “How can we best present the usage of a specialty computer lab? What is the data that can be collected from the lab and how can we present illustrations of the lab’s day-to-day use?” The first step was considering what information offered by a computer is most useful to anyone interested in seeing how a computer lab is being used. Generally, the list of processes running on a machine is a quick and easy way to gauge what’s being worked on.

To retrieve the specifics, software is written and installed onto each machine in Carpenter lab. This software periodically polls the computer’s processes to see how it is being used. This allows the setup to run without requiring anyone in one place at a specific time. Non-user specific data is then uploaded to the Raspberry Pi which interprets it and displays it on the web, and in turn, to a screen. Despite the Pi’s size (imagine a 2” thick credit card), it is capable of receiving the twelve sessions of input and publishing it to a website in a manner so that the information is easy to view.

This card-sized device can also fit behind a television screen and directly transmit what it finds. The information is displayed in a way to show comparisons to calendar dates (academic and otherwise) to understand trends and correlations for lab use. Once other methods of illustration and graphing are found, they’ll be offered as additional options of viewing. This implementation combines disciplines in both software and web based developments. This involves 12 independent computers accessing relevant data, formatting it for distribution across a network and sending it to a Raspberry Pi device. Although intended to display within the lab, the live findings can be accessed outside of the building using the website.