Optimizing Workflow Scheduling using Max-Min Algorithm in Cloud Environment (original) (raw)
Related papers
A New Modified Max-min Workflow Scheduling Algorithm for Cloud Environment
2019 15th International Conference on Electronics, Computer and Computation (ICECCO), 2019
In cloud computing, Service rendering is achieved through virtualization and the use of very large scalable resources where users access services by paying for the share of resources used by them. Several workflow scheduling techniques have been proposed for efficient scheduling. Max-min approach is prominent among them, its features includes efficiency and the ability to guarantee optimal or near optimal solution because of its heuristic nature. Several enhancement to its functionalities have been proposed in the literature and works are still ongoing for improving it. This paper proposed a solution by minimizing total execution time where workflow tasks length are invariably considered before proper scheduling. Our proposed work is compared with standard Max-Min scheduling scheme. The proposed scheme outperforms the standard scheme; two realistic scientific workflows were used as test data for the simulation.
SCIENTIFIC WORKFLOW SCHEDULING IN CLOUD COMPUTING ENVIRONMENT: A SURVEY
IAEME PUBLICATION, 2018
Workflow scheduling is one of the most challenging issues in cloud computing. Workflow is widely used paradigm in collaborative research and managing complex large scale distributed application. Various distributed environment such as cluster, grid and cloud use workflow to process complex and discrete tasks. Each task may include entering data, processing, accessing software, or storage functions. The task resource mapping, QoS requirement, on-demand resource provisioning, performance fluctuation and failure management in workflow scheduling is considered to be an NP-hard problem. An efficient scheduling algorithms are required to select the best suitable resources for workflow execution. In this paper, a comprehensive survey of workflow scheduling strategies that have been proposed for cloud computing platforms to help researchers systematically and objectively gather and aggregate research evidences.
Comparative analysis of Scientific Workflow Scheduling in Cloud Environment
International Conference on Innovations in Power and Advanced Computing Technologies [i-PACT2017], 2017
Cloud platform is the model of parallel and distributed computing. It offers the facilities by pay-as per-usage policy. Also provides the platform for high performance application like scientific applications. Virtualization, elasticity and pay by use are the important features of cloud computing. It allows the user to submit their tasks by providing varieties of resources through internet. Since cloud provides the facility of running multiple tasks simultaneously, an eminent scheduling algorithm is needed for the better performance. Numerous task based schedulers are available in the cloud environment. In this paper we have done the analysis of various traditional scheduling algorithms based on the parameters like average makespan and cost.
Critical Greedy Based Optimized Scheduling of Scientific Workflows in Cloud Computing
International Journal of Scientific and Engineering Research
Cloud computing is a model for enabling ubiquitous and ondemand network access to a shared pool of configurable computing resources that can be rapidly provisioned and released with minimal management effort or service provider interaction. Cloud computing is the latest distributed and utility computing paradigm and it differs tremendous opportunities to solve large scale scientific problems. Largescale scientific applications typically have to execute a large number of homogeneous and concurrent activities. Scientific applications can be modeled as workflows. Scheduling the tasks in the workflow is an NP-hard problem and in this work we focus on optimizing the scheduling process. One of the most challenging problems in Cloud computing is the workflow scheduling the problem of satisfying the Quality of Service of the users as well as minimizing the cost of workflows executions .We have to formulate a task scheduling problem to minimize the workflow end-to-end delay under a user-specified financial constraint. In this paper we have proposed Greedy-Based Algorithm in cloud computing to implement a new scheduling algorithm that can minimize the End to end delay in cloud environment.
The rise in demand for cloud resources (network, hardware and software) requires cost effective scientific workflow scheduling algorithm to reduce cost and balance load of all jobs evenly for a better system throughput. Getting multiple scientific workflows scheduled with a reduced makespan and cost in a dynamic cloud computing environment is an attractive research area which needs more attention. Scheduling multiple workflows with the standard Max-Min algorithm is a challenge because of the high priority given to task with maximum execution time first. To overcome this challenge, we proposed a new mechanism call Expanded Max-Min (Expa-Max-Min) algorithm to effectively give equal opportunity to both cloudlets with maximum and minimum execution time to be scheduled for a reduce cost and time. Expa-Max-Min algorithm first calculates the completion time of all the cloudlets in the cloudletList to find cloudlets with minimum and maximum execution time, then it sorts and queue the cloudlets in two queues based on their execution times. The algorithm first select a cloudlet from the cloudletList in the maximum execution time queue and assign it to a resource that produces minimum completion time, while executing cloudlets in the minimum execution time queue concurrently. The experimented results demonstrats that our proposed algorithm, Expa-Max-Min algorithm, is able to produce good quality solutions in terms of minimising average cost and makespan and able to balance loads than Max-Min and Min-Min algorithms.
Minimum Makespan Cloud Workflow Scheduling Algorithm for Scientific Workflows
— Cloud computing is a large-scale distributed computing that facilitates the execution of scientific and commercial applications by offering scalable, responsive and flexible computing platform. Scientific applications such as Montage, Cybershake and Epigenomics under the areas like astronomy, earth sciences and bioinformatics, are generally modeled as workflows. Scheduling workflow applications in dynamic and cost-driven environment like cloud is an interesting area for research. The key concern for the execution of performance driven applications, such as workflows, is to design a scheduling algorithm which chooses the suitable resource for workflow execution to minimize the makespan (execution time) of a workflow and to satisfy the user Quality of Service (QoS) requirements by using a set of heterogeneous resources available over the cloud. In this paper, Minimum Makespan Cloud Workflow Scheduling Algorithm (MMCWS) is proposed for addressing this challenge. MMCWS algorithm is designed to effectively optimize the overall execution time of workflow execution while reducing the cost. The MMCWS algorithm is evaluated using WorkflowSim for four real-world synthetic scientific workflow applications. The simulation results confirm that the proposed scheduling algorithm outperforms the related well-known scheduling approaches. Further, the performance comparison demonstrates its performance superiority over the existing methods.
Multi-priority scheduling algorithm for scientific workflows in cloud
Bulletin of Electrical Engineering and Informatics, 2024
The public cloud environment has emerged as a promising platform for executing scientific workflows. These executions involve leasing virtual machines (VMs) from public services for the duration of the workflow. The structure of the workflows significantly impacts the performance of any proposed scheduling approach. A task within a workflow cannot begin its execution before receiving all the required data from its preceding tasks. In this paper, we introduce a multi-priority scheduling approach for executing workflow tasks in the cloud. The key component of the proposed approach is a mechanism that logically orders and groups workflow tasks based on their data dependencies and locality. Using the proposed approach, the number of available VMs influences the number of groups (partitions) obtained. Based on the locality of each group’s tasks, the priority of each group is determined to reduce the overall execution delay and improve VM utilization. As the results demonstrate, the proposed approach achieves a significant reduction in both execution costs and time in most scenarios.
A survey on provisioning and scheduling algorithms for scientific workflows in cloud computing
THE 2ND UNIVERSITAS LAMPUNG INTERNATIONAL CONFERENCE ON SCIENCE, TECHNOLOGY, AND ENVIRONMENT (ULICoSTE) 2021
In the last decade, a scientific workflow has become the dominant trend to enable specialists to accelerate scientific developments in various areas. The scientific workflow requires high computing, therefore, cloud computing is exploited, as it provides the computing required for scientific workflow with a good cost. The frequent use of scientific workflows by researchers in cloud computing stimulates the need to find mechanisms to manage scientific workflow, providing resources, and scheduling in a way that achieves fairness for all users and meets the required Quality of Service (QoS). In a cloud computing environment, scheduling scientific workflows is an NP-hard problem. A lot of work has been done to resolve the decision problem of scheduling workflows, provision of resources, or both. In this survey, many papers related to scientific workflow algorithms of provisioning and scheduling in a cloud computing system are reviewed. We investigate and classify numerous related articles in the field. The objectives of each study, the application domains, and tools that were used are highlighted.
Compute-Intensive Workflow Scheduling in Multi-Cloud Environment
—Workflow scheduling is recognized as well-known NP-complete problem in the perspective of cloud computing environment. Workflow applications always need high compute-intensive operations because of the presence of precedence-constrains. The scheduling objective is to map the workflow application to the VMs pool at available cloud datacenters such that the overall processing time (makespan) is to be minimized and average cloud utilization is maximized. In this paper, we propose a two phase workflow scheduling algorithm with a new priority scheme. It considers the ratio of average communication cost to the average computation cost of the task node as a part of prioritization process in the first phase. Prioritized tasks are mapped to suitable virtual machines in the second phase. Proposed algorithm is capable of scheduling large size workflows in heterogeneous multi-cloud environment. The proposed algorithm is simulated rigorously on standard scientific workflows and simulated results are compared with the existing dependent task scheduling algorithms as per the assumed cloud model. The results remarkably show that the proposed algorithm supercedes the existing algorithms in terms of makespan, speed-up, schedule length ratio and average cloud utilization .
A Unified Mechanism for Cloud Scheduling of Scientific Workflows
IEEE Access, 2022
Scheduling plays a vital role in the efficient utilization of the available resources in clouds. This paper investigates the capabilities of the current scheduling algorithms of WorkFlowSim framework for processing scientific workflows. These investigations used four different sizes of workloads each, for, five well-known workflows. It was revealed that none of the existing algorithms is capable of efficiently executing all the four sizes of workload for the complete set of workflows. Different algorithms performed better, when they were applied to various workloads of a particular workflow. This fact was used in developing an improved unified mechanism, which is capable of using an existing algorithm that performed well in the past, against the given workload. Evaluation results showed that the proposed mechanism improved over the existing algorithms for 4 out of 5 workflows (Epigenomics, Inspiral, Cyber Shake, and Montage), when tested against an aggregated load of all sizes, in terms of simulation time. For the workflow named SIPHT, however, it responded exactly the same as Max-Min algorithm. The minimum and maximum improvements, against the existing best and worst algorithms, in percentage, for Epigenomics, Inspiral, SIPHT, Cyber Shake and Montage were 16-63, 30-68, 0-69, 30-68, and 9-71 in corresponding order. This work has an additional overhead in terms of a dedicated module to find and store algorithmic performance. It is, however, required once and, thus, the increase in execution time might be marginal. The future work intends to check the impact of compute time towards optimization parameters such as makespan, pricing and deadlines.