Conference Paper

Dynamic Task Scheduling using Genetic Algorithms for Heterogeneous Distributed Computing.

DOI: 10.1109/IPDPS.2005.184 Conference: 19th International Parallel and Distributed Processing Symposium (IPDPS 2005), CD-ROM / Abstracts Proceedings, 4-8 April 2005, Denver, CO, USA
Source: DBLP


An algorithm has been developed to dynamically sched- ule heterogeneous tasks on heterogeneous processors in a distributed system. The scheduler operates in an environ- ment with dynamically changing resources and adapts to variable system resources. It operates in a batch fashion and utilises a genetic algorithm to minimise the total exe- cution time. We have compared our scheduler to six other schedulers, three batch-mode and three immediate-mode schedulers. We have performed simulations with randomly generated task sets, using uniform, normal, and Poisson dis- tributions, whilst varying the communication overheads be- tween the clients and scheduler. We have achieved more effi- cient results than all other schedulers across a range of dif- ferent scenarios while scheduling 10,000 tasks on up to 50 heterogeneous processors.

Full-text preview

Available from:
  • Source
    • "A plethora of heuristics such as clustering algorithms (Palis et al. 1996, Topcuoglu et al. 2002), list scheduling algorithms (Augonnet et al. 2011, Topcuoglu et al. 1999), task duplication based algorithms (Hagras & Janeèek 2005, Park & Choe, 2001, Ranaweera & Agrawal, 2000), genetic algorithms (Oh&Wu, 2004, Ulusoy 2004), simulated annealing (Braun et al. 2001, Kazem et al. 2008, Wanneng & Shijue 2006), tabu search (Porto et al. 2000, Porto & Ribeiro 1995) and particle swam optimization (Jarboui et al. 2008, Salman et al. 2002) have been proposed in literature for the optimal solution of scheduling problem. Static (Shirazi et al. 1990) as well as dynamic scheduling (Page & Naughton 2005, Rotithor 1994) schemes are generally employed for the optimal solution. Static schemes augmented with different "
    [Show abstract] [Hide abstract]
    ABSTRACT: Heterogeneous Distributed Computing Systems (HeDCS) efficiently utilize the heterogeneity of diverse computational resources which are interlinked through high speed networks for executing a group of computing intensive applications. Directed acyclic graphs (DAGs) are usually used to represent these parallel applications with varied computational requirements and constraints. The optimal scheduling of the given set of precedence constrained tasks to available resources is a core concern in HeDCS and is known to be NP Complete problem. Non deterministic nature of application programs and heterogeneous environment are the main challenges in designing, implementing and analyzing phases of task scheduling techniques. A myriad of heuristic and meta-heuristic approaches have been proposed in the literature to solve this complex problem. The basic purpose of this study is to cover ANN based task scheduling strategies in the distributed computing environment perspective. Further existing scheduling heuristics could be classified in a new state of art classification including the description of frequently used parameters in the mentioned scheduling strategies. The flexible and powerful nature of ANN for identifying the data patterns, underlying time and other constraints and learning capabilities have shown to be a promising candidate among other heuristics.
    Full-text · Article · Dec 2015
  • Source
    • "Dhodhi et al. [13] also presented a GA for task scheduling on heterogeneous systems, where a new encoding has been used to represent the feasible solutions. It is remarkable that these GA-based multi-objective algorithms [10] [11] [12] [13] were implemented for the general tasks without considering time constraints. Yoo et al. [14] proposed a hybrid multi-objective algorithm based on GA and Simulated Annealing (SA) for scheduling of soft real-time tasks. "
    [Show abstract] [Hide abstract]
    ABSTRACT: Scheduling of real-time tasks in multiprocessor systems is a NP-hard problem. Recently, swarm intelligence algorithms have been efficiently applied for this problem. Real-time tasks can be classified into hard real-time tasks and soft real-time tasks. The aim of hard real-time task scheduling algorithms is to meet all tasks deadline constraints. However, slight violation is not critical, in the case of soft real-time tasks. In this paper, a new algorithm based on artificial bee colony (ABC) is proposed for scheduling of soft real-time tasks. In this method, a hybrid neighborhood search mechanism is introduced to improve the convergence of ABC. Experimental results demonstrate the effectiveness of proposed algorithm for scheduling of soft real-time tasks in heterogeneous multiprocessor systems.
    Full-text · Conference Paper · May 2014
  • Source
    • "In the case of Parallel machine scheduling, there are many literatures surrounding the multiobjective problem. The use of Holland's genetic algorithms [3] (GAs) in scheduling, which apply evolutionary strategies to allow for the fast exploration of the search space of schedules, allows good solutions to be found quickly and for the scheduler to be applied to more general problems [4]. E. Kim et al. [5] considered a deterministic scheduling problem where multiple jobs with s-precedence relations are processed on multiple identical parallel machines. "
    [Show abstract] [Hide abstract]
    ABSTRACT: Multiprocessor task scheduling is a NP-hard problem and Genetic Algorithm (GA) has been revealed as an excellent technique for finding an optimal solution. In the past, several methods have been considered for the solution of this problem based on GAs. But, all these methods consider single criteria and in the present work, minimization of the bi-criteria multiprocessor task scheduling problem has been considered which includes weighted sum of makespan & total completion time. Efficiency and effectiveness of genetic algorithm can be achieved by optimization of its different parameters such as crossover, mutation, crossover probability, selection function etc. The effects of GA parameters on minimization of bi-criteria fitness function and subsequent setting of parameters have been accomplished by central composite design (CCD) approach of response surface methodology (RSM) of Design of Experiments. The experiments have been performed with different levels of GA parameters and analysis of variance has been performed for significant parameters for minimisation of makespan and total completion time simultaneously.
    Full-text · Article · Jan 2014
Show more