We propose a parallel and distributed computation of genetic local search with irregular topology in distributed environments. The scheme we propose is implemented with tree network topologies where each computing element carries out genetic local search on its own chromosome set and communicates with its parent when the best solution of each generation is improved. We evaluate the proposed algorithm in a grid simulation environment implemented on a PC-cluster. We test our algorithm on four types topologies: star, line, balanced binary tree and sided binary tree, and find that the topology's depth and the number of independent search nodes influences on the evolution process. Furthermore, we observe in the experiment 'Reset' mechanism of the population after convergence is so useful in grid computing environments.