Call us Today ! +918886268863 | [email protected]

A New Double Rank-based Multi-workflow Scheduling with Multi-objective Optimization in Cloud Environments

A New Double Rank-based Multi-workflow Scheduling with Multi-objective Optimization in Cloud Environments

ABSTRACT:

Workflow scheduling in clouds has been extensively researched. Many workflows from different users could be submitted to clouds at the same time and cloud providers should handle them simultaneously. So, it is necessary to consider the problem of scheduling multi-workflow. In addition, cloud computing systems can offer some special features, like Pay-Per-Use and Quality of Service (QoS) over the Internet. The scheduler has to consider the tradeoffs between different QoS parameters in order to satisfy the QoS requirements. Hence, how to schedule multiple heterogeneous workflows in the meanwhile to balance multiple objectives is a big challenge. The majority of the existing multi-workflow scheduling algorithms are based on QoS constrained approaches and attempt to optimize one objective while taking other QoS factors as constraints. Meanwhile, most of the multi-objective optimization scheduling works aim to deal with single-workflow. Conversely, this paper focuses on QoS optimization approaches by finding trade-off schedules to execute multi-workflow on cloud computing resources so as to balance multi-objective. To this end, a new double rank-based task sequencing method is proposed and integrated with a multi-objective heuristic algorithm for multi-workflow scheduling. Different algorithms are evaluated using various well-known real-world workflows and simulated workflows. The performance evaluation results demonstrate that the proposed approach is capable of generating efficient schedules with high quality in terms of meeting multi-objective for multiple workflows.

INTRODUCTION:

The workflow scheduling problem has been extensively studied in the field of the cluster, grid, and cloud computing. Specifically, cloud environments has been attracting more and more attention for workflow execution because of its characteristics of on-demand access and pay-as-use. As cloud computing emerges, users can now have easy access to on-demand computing resources. Due to the cloud can serve many users simultaneously, it is common that many workflows submitted from different users need to be scheduled concurrently. In addition, the flexibility and ability to easily scale the number of cloud resources lead to a trade-off between the two conflicting time and cost requirements. Therefore, how to efficiently schedule concurrent workflows with multiobjective becomes an important issue in cloud environments. The scheduling approaches which consider multiple QoS indexes can be classified into QoS constrained schedulings and QoS optimization schedulings. The QoS constrained algorithms mainly include budget-constrained algorithms and deadline-constrained algorithms. Most of the existing works on scheduling multi-workflow, considering multiple QoS indexes, have followed the QoS constrained approach. In general, a QoS constrained approach attempts to assign a portion of the overall workflow QoS constraint to its individual tasks. Then, the sub-constraint that each task preserves decides the task scheduling order or is used to guide the resource allocation for objective optimization. A schedule is accepted if the constraint is met. In, the authors focused on multiworkflow scheduling to minimize makespan while meeting budget constraints. Two budget distributions, i.e., Fastest First Task-based Distribution (FFTD) and Slowest-First Taskbased Distribution (SFTD), were proposed to distribute budget to each task. A dynamic heuristic resource provisioning and scheduling algorithm was proposed to deal with multiworkflow scheduling. In, periodical workflow applications on cloud resources to minimize the total renting cost with deadline constraints were considered. A Precedence Treebased Enumeration Scheme (PTES) was developed to find one possible solution. Based on the PTES framework, three types of initial schedule construction methods and two improvement procedures were presented. In these processes, the time-related constraint was used for task selection and the cost-related factor was taken as the optimization objective. However, some limitations exist for the QoS constrained method. First, obtaining sub-QoS distribution is non-trivial due to the difficulty of getting sub-constraint for each task from the overall QoS constraint. Second, it is difficult to define constraint information such as a deadline constraint for the constrained optimization model since users do not know the range of the execution time in advance. Aiming to overcome the above-mentioned disadvantages of QoS-constrained scheduling. Some research tries to optimize multiple objectives simultaneously based on multi-objective QoS optimization approaches. Pareto-based trade-off solutions can be obtained by the approaches. Heuristic-based and metaheuristic-based are two kinds of methods to solve this kind of problem. For the multi-objective QoS optimization-based scheduling, most of the existing research focuses on a single workflow model. Multi-objective Heterogeneous Earliest Finish Time (MOHEFT) algorithm, developed, is the most typical heuristic-based multi-objective optimization method. Apart from MOHEFT-related methods, most of the multi-objective solutions are based on meta-heuristic algorithms. In, an evolutionary optimization-based algorithm was proposed to solve a single workflow scheduling problem with the objectives of both makespan and cost. In, it targeted to minimize makespan and cost while satisfying the reliability constraint based on particle swarm optimization technology. In [8], a co-evolutionary population-based ant colony system to deal with execution time and execution cost was proposed.

EXISTING SYSTEM:

Many workflows from different users could be submitted to clouds at the same time and cloud providers should handle them simultaneously. So, it is necessary to consider the problem of scheduling multi-workflow. In addition, cloud computing systems can offer some special features, like Pay-Per-Use and Quality of Service (QoS) over the Internet. The scheduler has to consider the tradeoffs between different QoS parameters in order to satisfy the QoS requirements. Hence, how to schedule multiple heterogeneous workflows in the meanwhile to balance multiple objectives is a big challenge. The majority of the existing multi-workflow scheduling algorithms are based on QoS constrained approaches and attempt to optimize one objective while taking other QoS factors as constraints.

DISADVANTAGES:

The majority of the existing multi-workflow scheduling algorithms are based on QoS constrained approaches and attempt to optimize one objective while taking other QoS factors as constraints.

PROPOSED SYSTEM:

In this paper, we present a problem of multi-workflow multiobjective scheduling. To solve this kind of scheduling problem, we first extend four multi-workflow single-objective scheduling methods to multi-workflow multi-objective scheduling methods by integrating them with the MOHEFT method. Then, a new double rank-based multi-workflow multi-objective scheduling method is proposed. The five methods are tested and compared using both the real-world workflows and nine sets of randomly generated workflows. The results show that the proposed double rank-based method outperforms four other approaches in most cases. It is demonstrated that the newly proposed method is promising to deal with the problems of multi-workflow multi-objective scheduling.

ADVANTAGES:

we only demonstrate the effectiveness of the new proposed method on the two optimization objectives.

SYSTEM ARCHITECTURE DIAGRAM:

CONCLUSION:

In this paper, we present a problem of multi-workflow multiobjective scheduling. To solve this kind of scheduling problem, we first extend four multi-workflow single-objective scheduling methods to multi-workflow multi-objective scheduling methods by integrating them with the MOHEFT method. Then, a new double rank-based multi-workflow multi-objective scheduling method is proposed. The five methods are tested and compared using both the real-world workflows and nine sets of randomly generated workflows. The results show that the proposed double rank-based method outperforms four other approaches in most cases. It is demonstrated that the newly proposed method is promising to deal with the problems of multi-workflow multi-objective scheduling. This paper is the first attempt to optimize pure multiobjective for multi-workflow. The limitation of this paper is that we only demonstrate the effectiveness of the new proposed method on the two optimization objectives. In the future, other objectives which can reflect the characteristics of multiworkflow models should be also considered. Particularly, how to make fair scheduling for each workflow in the workflow set while achieving the whole objectives should be researched. In addition, other multi-objective optimization algorithms should be considered for multi-workflow scheduling.

REFERENCE

[1] W. N. Chen, and J. Zhang, An ant colony optimization approach to a grid workflow scheduling problem with various QoS requirements. IEEE Transactions on Systems, Man, and Cybernetics, Part C, 39(1), pp.29-43, 2008.

[2] H. Topcuoglu, S. Hariri, and M. Y. Wu, Performance-effective and low-complexity task scheduling for heterogeneous computing. IEEE transactions on parallel and distributed systems, 13(3), pp.260-274, 2002.

[3] S. Abrishami, M. Naghibzadeh, and D. H. Epema, Deadline-constrained workflow scheduling algorithms for infrastructure as a service clouds. Future Generation Computer Systems, 29(1), pp.158-169, 2013.

[4] H. Muhammad Hafizhuddin, ”Budget-constrained Workflow Applications Scheduling in Workflow-as-a-Service Cloud Computing Environments.” PhD diss., the University of Melbourne, Australia, 2020.

[5] H. Arabnejad, and J. G. Barbosa, A budget constrained scheduling algorithm for workflow applications. Journal of grid computing, 12(4), pp.665-679, 2014.

[6] J. Yu, and R. Buyya, Scheduling scientific workflow applications with deadline and budget constraints using genetic algorithms. Scientific Programming, 14(3, 4), pp.217-230, 2006.

 [7] L. Chen, X. Li, and R. Ruiz, Resource renting for periodical cloud workflow applications. IEEE Transactions on Services Computing, 13(1), pp.130-143, 2020.

[8] Z. G. Chen, Z. H. Zhan, Y. Lin, Y. J. Gong, T. L. Gu, F. Zhao, et.al, Multiobjective cloud workflow scheduling: A multiple populations ant colony system approach. IEEE transactions on cybernetics, 49(8), pp.2912-2926, 2018.

 [9] M. A. Rodriguez and R. Buyya, “A Taxonomy and Survey on Scheduling Algorithms for Scientific Workflows in IaaS Cloud Computing Environments,” Concurrency and Computation: Practice and Experience, vol. 29, no. 8, pp.1-23, 2017.

 [10] J. J. Durillo, R. Prodan, H. M. Fard, MOHEFT: A multi-objective list-based method for workflow scheduling. IEEE 4th International Conference on Cloud Computing Technology and Science, pp.185-192, 2012.

March 17, 2022

0 responses on "A New Double Rank-based Multi-workflow Scheduling with Multi-objective Optimization in Cloud Environments"

Leave a Message

Template Design © VibeThemes. All rights reserved.