Masashi Oiso

Accelerating steady-state genetic algorithms based on CUDA architecture

Parallel processing using graphic processing units (GPUs) have attracted much research interest in recent years. Parallel computation can be applied to genetic algorithms (GAs) in terms of the processes of individuals in a population. This paper describes the implementation of GAs in the compute unified device architecture (CUDA) environment. CUDA is a general-purpose computation environment for GPUs. The major characteristic of this study is that a steady-state GA is implemented on a GPU based on concurrent kernel execution. The proposed implementation is evaluated through four test functions; we find that the proposed implementation method is 3.0–6.0 times faster than the corresponding CPU implementation.

Evaluation of Generation Alternation Models in Evolutionary Robotics

For efficient implementation of Evolutionary Algorithms (EA) to a desktop grid computing environment, we propose a new generation alternation model called Grid-Oriented-Deletion (GOD) based on comparison with the conventional techniques. In previous research, generation alternation models are generally evaluated by using test functions. However, their exploration performance on the real problems such as Evolutionary Robotics (ER) has not been made very clear yet. Therefore we investigate the relationship between the exploration performance of EA on an ER problem and its generation alternation model. We applied four generation alternation models to the Evolutionary Multi-Robotics (EMR), which is the package-pushing problem to investigate their exploration performance. The results show that GOD is more effective than the other conventional models.

Grid task scheduling algorithm R3Q for evolution strategies

A computational method for implementation of evolution strategies (ES) in grid computing environments is discussed. In this paper, list scheduling with round-robin order replication (RR) is adopted to reduce waiting times due to synchronization in ES. However, RR is suitable for coarsegrained tasks. For ES as medium-grained tasks, we propose a new technique to reduce the communication overhead, called the remote work queue (RWQ) method. We then define round- robin replication remote work queue (R3Q) as RWQ with RR. Our results show that R3Q can reduce both the synchronous waiting time and communication time, and provides efficient forced termination of tasks compared to other methods.

Application of Grid Task Scheduling Algorithm RR to Medium-Grained Evolution Strategies

A computational method for implementation of Evolution Strategies (ES) in Grid computing environments is discussed. In this paper, list scheduling with Round-robin order Replication (RR) is adopted to reduce waiting times due to synchronization in Medium-grained ES. Our results show that the replication in RR can reduce the synchronous waiting time in comparison with Work Queue (WQ) methods.

Application of grid task scheduling algorithm R3Q to evolutionary multi-robotics problem

A computational method for the implementation of an evolutionary multi-robotics (EMR) problem in grid computing environments is discussed. Due to the synchronization requirements of evolutionary algorithms (EAs), when the EMR problem is deployed in the grid environment there is a higher waiting time overhead because of medium-grained tasks. The round-robin replication remote work queue (R3Q) is adopted to reduce both the synchronous waiting time and communication time. In the current research, the performance of the grid scheduling is evaluated using uniform computational granularity despite that many problems such as EMR have nonuniform computational granularity. Therefore, in order to evaluate R3Q on nonuniform computational granularity, the cooperative object pushing EMR problem is adopted; and R3Q is compared with grid scheduling algorithms Work Queue (WQ), and list scheduling with round-robin order replication (RR). Our results show that R3Q is effective for tasks which have nonuniform computational granularity.

Desktop Grid Environment for Evolution Strategies using Grid Task Scheduling Algorithm RR

This paper empirically investigates the effect of Desktop Grid Environment on the computational cost of Evolution Strategies (ES). Especially, for task scheduling, List Scheduling with Round-robin order Replication (RR) is adopted to reduce waiting time due to synchronization in ES. Computer simulations are conducted which compare with the loss energy of theoretical maximum value and an actual maximum measurement on Desktop Grid Environment for ES. Results on RR prove that the loss energy of an actual maximum measurement is less than that of theoretical maximum value. Moreover, RR can reduce the synchronous waiting time and improve parallel efficiency in comparison with Work Queue algorithm.

Grid Task Scheduling Algorithm R3Q for Evolving Artificial Neural Networks

Task scheduling algorithms for evolving artificial neural networks (EANNs) in grid computing environments is discussed. In this paper, list scheduling with round-robin order replication (RR) is adopted to reduce waiting times due to synchronization. However, RR is suitable for coarse-grained tasks. For EANNs as medium-grained tasks, we propose a new technique to reduce the communication overhead, called the remote work queue (RWQ) method. We then define round-robin replication remote work queue (R3Q) as RWQ with RR. Our results show that R3Q can reduce both the synchronous waiting time and communication time %, and provides efficient forced termination of tasks compared to other methods.