SC is the International Conference for
 High Performnance Computing, Networking, Storage and Analysis

SCHEDULE: NOV 13-19, 2010

Adaptive Resolution Genetic Algorithm to Solve MINLPs over nVidia Fermi GPU using CUDA

SESSION: Research Poster Reception

EVENT TYPE: Poster

TIME: 5:15PM - 7:00PM

AUTHOR(S):Asim Munawar, Mohamed Wahib, Masaharu Munetomo, Kiyoshi Akama

ROOM:Main Lobby

ABSTRACT:
We propose an adaptive resolution approach to genetic algorithm (arGA) for solving Non convex Mixed Integer Non-Linear Programming (MINLP) problems. The technique uses the entropy measure of each variable to adjust the intensity of the genetic search around promising individuals. Performance is further improved by hybridization with adaptive resolution based local search (arLS) operator. In this poster, we describe the challenges and design choices involved in parallelization of this algorithm to solve complex MINLPs over nVidia Fermi GPU using CUDA. Results section shows several numerical tests and performance measurements obtained by running the algorithm over an nVidia Fermi GPU. We show that for difficult problems we can obtain a speedup of up to 15x with double precision and up to 34x with single precision.

Chair/Author Details:

Asim Munawar - Hokkaido University

Mohamed Wahib - Hokkaido University

Masaharu Munetomo - Hokkaido University

Kiyoshi Akama - Hokkaido University

Add to iCal  Click here to download .ics calendar file

Add to Outlook  Click here to download .vcs calendar file

Add to Google Calendarss  Click here to add event to your Google Calendar

   Sponsors    IEEE    ACM