Description
The goal of this project is to develop a distributed implementation of genetic programming on cloud environment that reduces response time by leveraging the vast computational and storage resources in a cloud-computing environment. The distributed genetic programming framework was modified and implemented on a MapReduce/Hadoop based cloud computing system. The performance evaluation of the system demonstrates that the accuracy is ensured while providing reduced response time and effective usage of computational and storage resources.
MapReduce is the distributed programming model used to speedup the implementation of the genetic algorithm in a Hadoop based cloud computing environment. The results show that speedup is achieved by dividing the input data into several blocks and to decrease the execution time. By dividing the input data, which is the candidate solutions population, forces Hadoop to handle theses data blocks by generating approximate number of Map functions.
Second job of feature extraction in distributed processing model
CUDA programming applied to constrained TSP using a genetic algorithm
Martim –
Outstanding work!