In 1993-94 my adviser asked me to run the theoretical component of the experiments our group was running in plasma chambers. He asked me to use his VisualBasic code because that way he could understand it. However, in order to do dust-plasma simulations you need to run many-body simulations and with the current code I calculated it would take a few years to generate enough data to complete my thesis. Fortunately, at the time, I was also the network/workstation administrator for the astronomy labs which had 40+ workstations networked via TCP/IP.  Each night after people left, I would reboot the workstations and have all but one run a C client that ran a distributed part of the simulation and report back to a central server written in C that was listening on a port for the data generated. A scratch web sever and basic client based on what I saw emerging out of high energy physics. 

This was the beginning of the world wide web, distributed services, etc. The end result was that within a week I had more than enough runs to complete my thesis and got a firm grounding on this new emerging technology soon to be called the world wide web.