Most algorithms developed for scheduling applications on global Grids focus on a single Quality of Service(QoS) parameter such as execution time, cost or total data transmission time. However, if we consider more than one QoS parameter (e.g. execution cost and time, which may be in conflict) then the problem becomes more challenging. To handle such scenarios, it is convenient to use heuristics rather than a deterministic algorithm. In this paper, we have proposed a workflow execution planning approach using Multiobjective Differential Evolution (MODE). Our goal was to generate a set of trade-off schedules according to two user specified QoS requirements (time and cost), which will offer more flexibility to users when estimating their QoS requirements. We have compared our results with a well-known baseline algorithm 'Pareto-archived Evolutionary Strategy (PAES)'. Simulation results show that the modified MODE is able to find significantly better spread of compromise solutions compared with that of PAES.The application of nature's heuristics for scheduling of applications has been explored by various researches [9][10][11][12]. There are numerous studies reporting work done on scheduling Directed Acyclic Graph (DAG)-based task graphs in multiprocessor systems [5]. Genetic algorithms and Heterogeneous Earliest Finish Time have been extended by the ASKALON project [13,14] to schedule scientific applications in Grid environments. Recently, in [15], a different base-line algorithm for Multiobjective Optimization was tested on different workflow models.Evolutionary techniques are now a widely used approach for tackling complex multiobjective optimization problems. We find numerous examples of them in the literature. For example, in [16] the authors proposed a local-search-based enhancements for evolutionary computation. Similar approaches can be found in [17][18][19][20]. Evolutionary and genetic-based optimization techniques are also found in other Grid-oriented applications (especially in Data-Grid applications), for review, readers are encouraged to refer to [21,22]. In a similar study [23], the proposed model was compared with Max-Min and Min-Min heuristics described in [4]. Although most of the techniques 'Delete-Shift-Insert' operations). For simplicity, we have chosen F=1. However, in the case of general DE, each of the variables is real numbers and the fitness is measured in terms of mathematical operations on those real numbers. But in our case, each of these variables (characters/numbers) refers to different services, hence, we cannot map these numbers/characters to real numbers and apply mathematical operations as in [7,8]. Consequently, in place of adding the difference with each value of P i1 , we mutate P i1 mat . I.e., we mutated D number of genes in P i1 where D is the Ulam distance measure described as above. The process is done in Algorithm 3, step 3. As result of this approach, we have just replaced the operation in Equation (13) with Algorithm 1. We can now devise the overall MODE from Algorithms 1 ...
One of the major difficulties when applying Multiobjective Evolutionary Algorithms (MOEA) to real world problems is the large number of objective function evaluations. Approximate (or surrogate) methods offer the possibility of reducing the number of evaluations, without reducing solution quality. Artificial Neural Network (ANN) based models are one approach that have been used to approximate the future front from the current available fronts with acceptable accuracy levels. However, the associated computational costs limit their effectiveness. In this work, we introduce a simple approach that has comparatively smaller computational cost and we have developed this model as a variation operator that can be used in any kind of multiobjective optimizer. When designing this model, we have considered the whole search procedure as a dynamic system that takes available objective values in current front as input and generates approximated design variables for the next front as output. Initial simulation experiments have produced encouraging results in comparison to NSGA-II. Our motivation was to increase the speed of the hosting optimizer. We have compared the performance of the algorithm with respect to the total number of function evaluation and Hypervolume metric. This variation operator has worst case complexity of O(nkN 3 ), where N is the population size, n and k is the number of design variables and objectives respectively.
Most algorithms developed for scheduling applications on global Grids focus on a single Quality of Service (QoS) parameter such as execution time, cost or total data transmission time. However, if we consider more than one QoS parameter (eg. execution cost and time may be in conflict) then the problem becomes more challenging. To handle such scenarios, it is convenient to use heuristics rather than a deterministic algorithm. In this paper we have proposed a workflow execution planning approach using Multiobjective Differential Evolution (MODE). Our goal was to generate a set of trade-off schedules according to two user specified QoS requirements (time and cost). The alternative tradeoff solutions offer more flexibility to users when estimating their QoS requirements of workflow executions. We have compared our results with two baseline multiobjective evolutionary algorithms. Simulation results show that our modified MODE is able to find a comparatively better spread of compromise solutions.
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.
customersupport@researchsolutions.com
10624 S. Eastern Ave., Ste. A-614
Henderson, NV 89052, USA
This site is protected by reCAPTCHA and the Google Privacy Policy and Terms of Service apply.
Copyright © 2024 scite LLC. All rights reserved.
Made with 💙 for researchers
Part of the Research Solutions Family.