Monday, March 9, 2009

Current plan

We decided that a good way to start would be to pick some simple, embarrassingly parallel problem to try and make concurrent with metaprogramming techniques. So at the moment, I'm working on writing a simple Mandelbrot set computation in Ruby. At Prof. Frens advice, instead of actually drawing the image, I'll probably just output a .PPM file. To keep it even simpler, I'll just keep it monochrome.

From there, I'll take the time to familiarize myself with threading in Ruby and I'll make a variation of the algorithm that runs concurrently in the traditional manner for style and speed comparison with my end result. I'm hoping to be done with that before the end of next week (which should be easily doable if I can keep on task). From there, I'll be on spring break, which I'll spend holed up in Minneapolis ramming my head into metaprogramming, hopefully emerging with something that works for this problem. Having some of the best Thai food I've ever had only half a mile away should help.

No comments: