System S: Describing State in Functional Languages

I invented a mathematical system for analyzing programs (System S). I used System S to automatically find parts of programs which we can speed up by running them at the same time. On our new multicore computers this gives large performance increases, at best even doubling the speed with two cores. Currently, programs must be tediously re-written by hand to use multiple cores, but with my system, we can automatically and efficiently convert them to exploit the full power of our new multicore hardware. I have implemented and tested my system and it works well on multicore machines. System S also has applications in theoretical computer science (type theory) and programming language semantics (new evaluation strategies).

Name: Kevin Michael Ellis
Stats: 17 year-old Junior from Catlin Gabel in Portland, OR
Influential Person: Andrew Merrill, my computer science teacher
Career Goal: Go to a university and then grad school; do research in computer science
Field Interest: I read the book “The Structure and Interpretation of Computer Programs,” which got me interested the topics that lead to this project. I then started work on parallel programming because it is very important right now in computer science.

Tags: education, innovation, inspire, Intel, Intel International Science and Engineering Fair, Intel ISEF

This entry was posted on Friday, May 29th, 2009 at 3:47 pm and is filed under Uncategorized. You can follow any responses to this entry through the RSS 2.0 feed. You can leave a response, or trackback from your own site.