http://www.cs.newpaltz.edu/~lik/publications/Yuming-Xu-IS-2014.pdf

A genetic algorithm for task scheduling on heterogeneous computing systems using multiple priority queues

Lévy flight distribution: A new metaheuristic algorithm for solving engineering optimization problems

An improved genetic algorithm for task scheduling in the cloud environments using the priority queues: Formal verification, simulation, and statistical testing

Display Omitted Ensemble of particle swarm optimization algorithms with self-adaptive mechanism called EPSO is proposed in this paper.In EPSO, the population is divided into small and large subpopulations to enhance population diversity.In small subpopulation, comprehensive learning PSO (CLPSO) is used to preserve the population diversity.In large subpopulation, inertia weight PSO, CLPSO, FDR-PSO, HPSO-TVAC and LIPS are hybridized together as an ensemble approach.Self-adaptive mechanism is employed to identify the best algorithm by learning from their previous experiences so that best-performing algorithm is assigned to individuals in the large subpopulation. According to the No Free Lunch (NFL) theorem, there is no single optimization algorithm to solve every problem effectively and efficiently. Different algorithms possess capabilities for solving different types of optimization problems. It is difficult to predict the best algorithm for every optimization problem. However, the ensemble of different optimization algorithms could be a potential solution and more efficient than using one single algorithm for solving complex problems. Inspired by this, we propose an ensemble of different particle swarm optimization algorithms called the ensemble particle swarm optimizer (EPSO) to solve real-parameter optimization problems. In each generation, a self-adaptive scheme is employed to identify the top algorithms by learning from their previous experiences in generating promising solutions. Consequently, the best-performing algorithm can be determined adaptively for each generation and assigned to individuals in the population. The performance of the proposed ensemble particle swarm optimization algorithm is evaluated using the CEC2005 real-parameter optimization benchmark problems and compared with each individual algorithm and other state-of-the-art optimization algorithms to show the superiority of the proposed ensemble particle swarm optimization (EPSO) algorithm.

Ensemble particle swarm optimizer

Mobile Edge Computing (MEC) is considered an essential future service for the implementation of 5G networks and the Internet of Things, as it is the best method of delivering computation and communication resources to mobile devices. It is based on the connection of the users to servers located on the edge of the network, which is especially relevant for real-time applications that demand minimal latency. In order to guarantee a resource-efficient MEC (which, for example, could mean improved Quality of Service for users or lower costs for service providers), it is important to consider certain aspects of the service model, such as where to offload the tasks generated by the devices, how many resources to allocate to each user (specially in the wired or wireless device-server communication) and how to handle inter-server communication. However, in the MEC scenarios with many and varied users, servers and applications, these problems are characterized by parameters with exceedingly high levels of dimensionality, resulting in too much data to be processed and complicating the task of finding efficient configurations. This will be particularly troublesome when 5G networks and Internet of Things roll out, with their massive amounts of devices. To address this concern, the best solution is to utilize Machine Learning (ML) algorithms, which enable the computer to draw conclusions and make predictions based on existing data without human supervision, leading to quick near-optimal solutions even in problems with high dimensionality. Indeed, in scenarios with too much data and too many parameters, ML algorithms are often the only feasible alternative. In this paper, a comprehensive survey on the use of ML in MEC systems is provided, offering an insight into the current progress of this research area. Furthermore, helpful guidance is supplied by pointing out which MEC challenges can be solved by ML solutions, what are the current trending algorithms in frontier ML research and how they could be used in MEC. These pieces of information should prove fundamental in encouraging future research that combines ML and MEC.

Machine Learning Meets Computation and Communication Control in Evolving Edge and Cloud: Challenges and Future Perspective

Scheduling for directed acyclic graph (DAG) tasks with the objective of minimizing makespan has become an important problem in a variety of applications on heterogeneous computing platforms, which involves making decisions about the execution order of tasks and task-to-processor mapping. Recently, the chemical reaction optimization (CRO) method has proved to be very effective in many fields. In this paper, an improved hybrid version of the CRO method called HCRO (hybrid CRO) is developed for solving the DAG-based task scheduling problem. In HCRO, the CRO method is integrated with the novel heuristic approaches, and a new selection strategy is proposed. More specifically, the following contributions are made in this paper. (1) A Gaussian random walk approach is proposed to search for optimal local candidate solutions. (2) A left or right rotating shift method based on the theory of maximum Hamming distance is used to guarantee that our HCRO algorithm can escape from local optima. (3) A novel selection strategy based on the normal distribution and a pseudo-random shuffle approach are developed to keep the molecular diversity. Moreover, an exclusive-OR (XOR) operator between two strings is introduced to reduce the chance of cloning before new molecules are generated. Both simulation and real-life experiments have been conducted in this paper to verify the effectiveness of HCRO. The results show that the HCRO algorithm schedules the DAG tasks much better than the existing algorithms in terms of makespan and speed of convergence.

/pdf/a-hybrid-chemical-reaction-optimization-scheme-for-task-35bq0upa3g.pdf

A Hybrid Chemical Reaction Optimization Scheme for Task Scheduling on Heterogeneous Computing Systems

The 0-1 knapsack problem (KP01) is a well-known combinatorial optimization problem. It is an NP-hard problem which plays important roles in computing theory and in many real life applications. Chemical reaction optimization (CRO) is a new optimization framework, inspired by the nature of chemical reactions. CRO has demonstrated excellent performance in solving many engineering problems such as the quadratic assignment problem, neural network training, multimodal continuous problems, etc. This paper proposes a new chemical reaction optimization with greedy strategy algorithm (CROG) to solve KP01. The paper also explains the operator design and parameter turning methods for CROG. A new repair function integrating a greedy strategy and random selection is used to repair the infeasible solutions. The experimental results have proven the superior performance of CROG compared to genetic algorithm (GA), ant colony optimization (ACO) and quantum-inspired evolutionary algorithm (QEA).

Chemical reaction optimization with greedy strategy for the 0-1 knapsack problem

A DAG scheduling scheme on heterogeneous computing systems using double molecular structure-based chemical reaction optimization

http://www.cs.newpaltz.edu/~lik/publications/Aijia-Ouyang-CF-2015.pdf

Parallel hybrid PSO with CUDA for lD heat conduction equation

On the distributed or parallel heterogeneous computing systems, an application is usually decomposed into several independent and/or interdependent sets of cooperating subtasks and assigned to a set of available processors for execution. Heuristic-based task scheduling algorithms consist of the two typical phases of task prioritization and processor selection. However, heuristic-based task scheduling algorithms produce a different makespan (completion time /schedule length) using the different task prioritization on a distributed or parallel heterogeneous computing systems. Therefore, the role of a good scheduling algorithm is to efficiently assign each subtask to a priority depending on the resources needed to minimize makespan. In this paper, a multiple priority queueing genetic algorithm (MPQGA) for task scheduling on heterogeneous computing systems is proposed. The basic idea of our approach is to exploit the advantages of both evolutionary and heuristic based algorithms while avoiding their drawbacks. Our algorithm incorporates a genetic algorithm (GA) approach to assign priority for each subtask while using a heuristic based heterogeneous earliest finish time (HEFT) approach to search for a solution for mapping subtasks to processors. The software simulation results, over a large set of randomly generated graphs as well as graphs for real-world problems with various characteristics, show that the makespan is increased when the number of nodes or communication to computation ratios (CCR) increased and decreased with the increasing parallelism or number of available processors. The proposed MPQGA algorithm significantly outperforms several related algorithms in terms of the schedule quality. The average makespan reduction is about 5.3 %.

Yuming Xu

Papers

A Hybrid Chemical Reaction Optimization Scheme for Task Scheduling on Heterogeneous Computing Systems

Chemical reaction optimization with greedy strategy for the 0-1 knapsack problem

A DAG scheduling scheme on heterogeneous computing systems using double molecular structure-based chemical reaction optimization

Parallel hybrid PSO with CUDA for lD heat conduction equation

A Multiple Priority Queueing Genetic Algorithm for Task Scheduling on Heterogeneous Computing Systems