That's NAK in the upper left of the above fisheye view. Click on the above animated GIF image to see a much higher quality ~8MB time-lapse movie of the build party.
Like KLAT2 and KASY0 before it, we literally built NAK in a day... well, except for wiring the network. The build is different from earlier builds in that it is really a "re-build" party -- there was a lot of disassembly to be done as well as assembly.
The hardcopy call for helpers is here as a PDF. There was no prior experience required to help build NAK. We simply asked that people sign-up and be at 672 F. Paul Anderson Tower for the timeslots they indicated. We provided food and refreshments to anyone helping to construct the machine. The people who actually did sign-up or are otherwise known to have helped (in alphabetical order) are:
Bill Dieter (former UK faculty co-leader of KAOS)
Tim Mattox (the KAOS PhD graduate who created KASY0's SFNN)
In case you were wondering, the reason that we openly invite people to help is not really because we need so much help... we easily could build the machine ourselves over a few days. The goal is to give more people, especially students, the experience of having helped to build a cluster supercomputer. Our group develops the systems technology for cluster supercomputing; the more people we can show how to apply these technologies, the better.
The only thing set in stone is our name.