Differential Evolution Algorithm For Parallel Implementation,

Traditional parallel computing tasks are often a large parallel computer to complete, thus parallel research on parallel computer has become the main direction of research. With the development of economy and science, the computational tasks become more and more in the fields of biomedicine, weather forecasting and high-power physics. Their specialties are:lots of computational data, complicated algorithm, and large-scaled, most difficult computation. So, Computer's reliability and availability are becoming more and more important. Though a single computer's capability and reliability are becoming better and better, But a single computer can't complete some requests still. Therefore, people think of a method to build a high performance system with common PCs which are connected to each other by some ways, the system is called cluster. Cluster is the best substitute of expensive super computer, because it is very cheap and convenient, moreover, it demands very low for software and hardware. Cluster technology is a technology of organizing several computers to work together to make a more powerful computer. The technology can provide more availability and more reliability.Differential evolution algorithm was proposed by Storn R and Price K in 1995, it's a new simple and effective computing technology, but it was be paid attention to until 2000 by scholars of various countries, and have achieved some results. Its main features are simple, fast convergence, high stability and suitable for solving complex optimization problems. But there are also many areas for improvement, both in theory and in practice, the differential evolution algorithm is not currently ripe, so it is necessary to study to expand the application field, and use it to solve more problems.This paper analyzes the research status of parallel computing, parallel algorithms and differential evolution algorithm firstly, and then introduces the principle of differential evolution algorithm, do a more systematic study on algorithm-related issues, especially population size, mutation operator, crossover operator influence on the differential evolution algorithm. Build a parallel computing system and achieve parallel differential evolution algorithm is the focus of this article, we use the ring topology rather than the bus topology, so that each sub-group of the best individual will be based on ring topology, emigrated to the next sub-group, meanwhile two parallel model of the differential evolution algorithm was given in this paper and the second model is programmed. The results show that after the differential evolution algorithm being paralleled, the running time was reduced and the operational performance was also improved significantly, exchange of information sub-group assigned to different nodes of the scope of the performance of the algorithm has a significant impact. In addition, not all of the mutations differential evolution strategies are equally sensitive to constant migration.