[HiPerCoPS] A New Direction

I'm nearly finished with my two's complement multiplier. For a long time, I couldn't figure out why multiplication of negative numbers didn't work. On Tuesday, though, I sat down and fleshed it out by walking through the algorithm step-by-step. This means that I'm almost ready to move on to some new things. I need to implement an unsigned multiplier, and I might even take a shot at the hierarchical multiplier. Once those are all finished, the multiplier corner of this research will be pretty much exhausted. Dr. Delgado has suggested that I might be able to write a paper on what I've done so far, but we need to find a journal that is appropriate for this topic.

With the multiplier business coming to an end, it's time to start moving in a new direction. In my fault tolerance class we're required to do a term project, so I'd like to kill two birds with one stone by working on adding some kind of fault tolerance to the Washington architecture. I'm not really sure what shape this project will take yet, but I'll figure it out over the next few days.