Parallelizing Genetic Algorithms with GPGPU
نویسنده
چکیده
GPGPU has proved effective in speeding up many applications, notably those that exhibit “embarrassing” parallelism (vector and matrix arithmetic, graphics, image processing, etc.). Other applications have proved more challenging. In particular, little research has been published on GPGPU parallelization of genetic algorithms. Genetic algorithms are inherently sequential in nature, but there is still significant parallelization potential within each generation. This research presents a framework for GPGPU parallel processing of genetic algorithms, and demonstrates that significant speedups are readily achievable.
منابع مشابه
A New Approach to Solve N-Queen Problem with Parallel Genetic Algorithm
Over the past few decades great efforts were made to solve uncertain hybrid optimization problems. The n-Queen problem is one of such problems that many solutions have been proposed for. The traditional methods to solve this problem are exponential in terms of runtime and are not acceptable in terms of space and memory complexity. In this study, parallel genetic algorithms are proposed to solve...
متن کاملA Comprehensive Survey on Various Evolutionary Algorithms on GPU
This paper presents a comprehensive survey on parallelizing computations involved in optimization problem on Graphics Processing Unit (GPU) using CUDA (Compute Unified Design Architecture). GPU have multithread cores with high memory bandwidth which allow for greater ease of use and also more radially support a layer body of applications. Many researchers have reported significant speedups with...
متن کاملAcceleration of the Retinex algorithm for image restoration by GPGPU/CUDA
Retinex is an image restoration method that can restore the image’s original appearance. The Retinex algorithm utilizes a Gaussian blur convolution with large kernel size to compute the center/surround information. Then a log-domain processing between the original image and the center/surround information is performed pixel-wise. The final step of the Retinex algorithm is to normalize the resul...
متن کاملLossless data compression on GPGPU architectures
Modern graphics processors provide exceptional computational power, but only for certain computational models. While they have revolutionized computation in many fields, compression has been largely unnaffected. This paper aims to explain the current issues and possibilities in GPGPU compression. This is done by a high level overview of the GPGPU computational model in the context of compressio...
متن کاملParallelizing simulated annealing algorithms based on high-performance computer
We implemented five conversions of simulated annealing (SA) algorithm from sequential-to-parallel forms on high-performance computers and applied them to a set of standard function optimization problems in order to test their performances. According to the experimental results, we eventually found that the traditional approach to parallelizing simulated annealing, namely, parallelizing moves in...
متن کامل