Automatic Program Grading System from NTU

I can't believe that I forgot to mention this in my blog post about the training sessions we held in Taiwan the first week in March 2011. I need to remember to consult my notes when I'm writing up these reports.

Professor Pangfeng Liu from the Department of Computer Science and Information Engineering at National Taiwan University is conducting a parallel programming course this semester (Spring 2011). He is using a Google Sites page for the online distribution of course syllabus, worksheets and other materials. (At the time of this writing, the current course was located here.) The site even features a countdown to the midterm.

However, the coolest part of Dr. Liu's presentation was that the CSIE department at NTU have an automatic judging system for programming homework. The name of the system is Judge Girl. Judge Girl even has a fan page on Facebook. I don't have any more details than this, since I can't read Mandarin and I didn't put much more in my notes. Even so, just the notion that someone had created a system able to check and grade programming homework was enough to get me excited and interested.

My first thought was to wonder if such a system could be used to build, run and grade entries in the Intel Threading Challenge. The work and time savings to judges would be most welcome. We could even put up some judging data sets and let contestants run their codes against them on the judging platforms to determine if their entries are able to get the right answers and if their efforts are fast enough. *sigh*

When I was a grad TA back at Washington State University we had tried to develop something like this. There were many problems with the current state of technology and the university computing environment at that time that prevented us from creating a successful system. So, kudos to the NTU CSIE department for showing me that such a system is not only possible, but very feasible.
Para obtener información más completa sobre las optimizaciones del compilador, consulte nuestro Aviso de optimización.