scholarly journals Bi-objective scheduling algorithm for scientific workflows on cloud computing platform with makespan and monetary cost minimization approach

2021 ◽  
Author(s):  
Mirsaeid Hosseini Shirvani ◽  
Reza Noorian Talouki
Keyword(s):  
Cost Minimization ◽  
Scheduling Algorithm ◽  
Scientific Workflow ◽  
Scientific Workflows ◽  
Discrete Problem ◽  
Hybrid Cloud ◽  
Monetary Cost ◽  
Computing Platform ◽  
Cloud Architecture ◽  
Private And Public

AbstractScheduling of scientific workflows on hybrid cloud architecture, which contains private and public clouds, is a challenging task because schedulers should be aware of task inter-dependencies, underlying heterogeneity, cost diversity, and virtual machine (VM) variable configurations during the scheduling process. On the one side, reaching a minimum total execution time or makespan is a favorable issue for users whereas the cost of utilizing quicker VMs may lead to conflict with their budget on the other side. Existing works in the literature scarcely consider VM’s monetary cost in the scheduling process but mainly focus on makespan. Therefore, in this paper, the problem of scientific workflow scheduling running on hybrid cloud architecture is formulated to a bi-objective optimization problem with makespan and monetary cost minimization viewpoint. To address this combinatorial discrete problem, this paper presents a hybrid bi-objective optimization based on simulated annealing and task duplication algorithms (BOSA-TDA) that exploits two important heuristics heterogeneous earliest finish time (HEFT) and duplication techniques to improve canonical SA. The extensive simulation results reported of running different well-known scientific workflows such as LIGO, SIPHT, Cybershake, Montage, and Epigenomics demonstrate that proposed BOSA-TDA has the amount of 12.5%, 14.5%, 17%, 13.5%, and 18.5% average improvement against other existing approaches in terms of makespan, monetary cost, speed up, SLR, and efficiency metrics, respectively.

Cost Effective Heuristic workflow scheduling algorithm in Cloud under Deadline Constraint

2019 ◽  
Vol 12 ◽  
Author(s):  
Jasraj Meena ◽  
Manu Vardhan
Keyword(s):  
Cloud Computing ◽  
Scheduling Algorithm ◽  
Cost Effective ◽  
Scientific Workflow ◽  
Scientific Workflows ◽  
Workflow Scheduling ◽  
Performance Variation ◽  
Acquisition Delay ◽  
Very High ◽  
Deadline Constraint

Cloud computing is used to deliver IT resources over the internet. Due to the popularity of cloud computing, nowadays, most of the scientific workflows are shifted towards this environment. There are lots of algorithms has been proposed in the literature to schedule scientific workflows in the cloud, but their execution cost is very high as well as they are not meeting the user-defined deadline constraint. This paper focuses on satisfying the userdefined deadline of a scientific workflow while minimizing the total execution cost. So, to achieve this, we have proposed a Cost-Effective under Deadline (CEuD) constraint workflow scheduling algorithm. The proposed CEuD algorithm considers all the essential features of Cloud and resolves the major issues such as performance variation, and acquisition delay. We have compared the proposed CEuD algorithm with the existing literature algorithms for scientific workflows (i.e., Montage, Epigenomics, and CyberShake) and getting better results for minimizing the overall execution cost of the workflow while satisfying the user-defined deadline.

scholarly journals Fuzzy Theory-Based Data Placement for Scientific Workflows in Hybrid Cloud Environments

2020 ◽  
Vol 2020 ◽  
pp. 1-13
Author(s):  
Zheyi Chen ◽  
Xu Zhao ◽  
Bing Lin
Keyword(s):  
Data Transmission ◽  
Data Centers ◽  
Fuzzy Theory ◽  
Data Placement ◽  
Scientific Workflow ◽  
Scientific Workflows ◽  
Transmission Time ◽  
Fuzzy Data ◽  
Hybrid Cloud ◽  
Cloud Environments

In hybrid cloud environments, reasonable data placement strategies are critical to the efficient execution of scientific workflows. Due to various loads, bandwidth fluctuations, and network congestions between different data centers as well as the dynamics of hybrid cloud environments, the data transmission time is uncertain. Thus, it poses huge challenges to the efficient data placement for scientific workflows. However, most of the traditional solutions for data placement focus on deterministic cloud environments, which lead to the excessive data transmission time of scientific workflows. To address this problem, we propose an adaptive discrete particle swarm optimization algorithm based on the fuzzy theory and genetic algorithm operators (DPSO-FGA) to minimize the fuzzy data transmission time of scientific workflows. The DPSO-FGA can rationally place the scientific workflow data while meeting the requirements of data privacy and the capacity limitations of data centers. Simulation results show that the DPSO-FGA can effectively reduce the fuzzy data transmission time of scientific workflows in hybrid cloud environments.

scholarly journals Design of Energy Aware Scheduling Algorithm for Executing Scientific Workflows in Cloud

2019 ◽  
Vol 9 (2) ◽  
pp. 1305-1311
Keyword(s):  
Energy Efficient ◽  
Virtual Machines ◽  
Scheduling Algorithm ◽  
Scientific Workflow ◽  
Scientific Workflows ◽  
Cloud Environment ◽  
Scientific Applications ◽  
Energy Aware ◽  
Efficient Manner ◽  
The Given

The usage of cloud computing and its resources for the execution of scientific workflow is a rapidly increasing demand. The Scientific applications are generally large in scale; even a single scientific workflow includes more number of complex tasks. Execution of these tasks can be made successful only by deploying it in the cloud virtual machines, because only cloud environment can only provide very large number of computing assets. In cloud, every processing resource is given as Virtual Machine. Any scientific workflow deployed in the cloud needs large number of virtual machines so; huge amount of computational energy is spent by the virtual machines to execute multifaceted scientific workflows. Hence there arises the need to utilize the cloud resources in an energy efficient way. Also, if the virtual machines are planned to schedule in an energy efficient manner there is an increase of makepsan of the workflow which is going to be an important parameter for completing the workflow within the deadline. So, the need for executing scientific workflows in energy efficient way with reduced makespan becomes a major issue among the researchers. It also becomes very challenging task to executing a scientific workflow in within the given deadline of a task in the given workflow. To address these issues, a new Energy Aware workflow scheduling algorithm is proposed and designed with improved makespan for the execution of different scientific applications in cloud environment.

scholarly journals Monitoring of Grid Scientific Workflows

2008 ◽  
Vol 16 (2-3) ◽  
pp. 205-216
Author(s):  
Bartosz Balis ◽  
Marian Bubak ◽  
Bartłomiej Łabno
Keyword(s):  
Real Life ◽  
Scientific Workflow ◽  
Point Of View ◽  
Scientific Workflows ◽  
Monitoring Data ◽  
Monitoring Point ◽  
Loosely Coupled ◽  
Provenance Data ◽  
On Line ◽  
Monitoring Architecture

Scientific workflows are a means of conducting in silico experiments in modern computing infrastructures for e-Science, often built on top of Grids. Monitoring of Grid scientific workflows is essential not only for performance analysis but also to collect provenance data and gather feedback useful in future decisions, e.g., related to optimization of resource usage. In this paper, basic problems related to monitoring of Grid scientific workflows are discussed. Being highly distributed, loosely coupled in space and time, heterogeneous, and heavily using legacy codes, workflows are exceptionally challenging from the monitoring point of view. We propose a Grid monitoring architecture for scientific workflows. Monitoring data correlation problem is described and an algorithm for on-line distributed collection of monitoring data is proposed. We demonstrate a prototype implementation of the proposed workflow monitoring architecture, the GEMINI monitoring system, and its use for monitoring of a real-life scientific workflow.

scholarly journals A cost efficient model for minimizing energy consumption and processing time for IoT tasks in a Mobile Edge Computing environment

2020 ◽  
Author(s):  
João Luiz Grave Gross ◽  
Cláudio Fernando Fernando Resin Geyer
Keyword(s):  
Energy Consumption ◽  
Cost Minimization ◽  
Cost Model ◽  
Scheduling Algorithm ◽  
Edge Computing ◽  
Mobile Edge Computing ◽  
Total Energy Consumption ◽  
Data Intensive ◽  
Battery Capacity ◽  
Cost Efficient

In a scenario with increasingly mobile devices connected to the Internet, data-intensive applications and energy consumption limited by battery capacity, we propose a cost minimization model for IoT devices in a Mobile Edge Computing (MEC) architecture with the main objective of reducing total energy consumption and total elapsed times from task creation to conclusion. The cost model is implemented using the TEMS (Time and Energy Minimization Scheduler) scheduling algorithm and validated with simulation. The results show that it is possible to reduce the energy consumed in the system by up to 51.61% and the total elapsed time by up to 86.65% in the simulated cases with the parameters and characteristics defined in each experiment.

A Data Placement Strategy for Scientific Workflow in Hybrid Cloud

2018 ◽  
Author(s):  
Zhanghui Liu ◽  
Tao Xiang ◽  
Bing Lin ◽  
Xinshu Ye ◽  
Haijiang Wang ◽  
...  
Keyword(s):  
Data Placement ◽  
Scientific Workflow ◽  
Hybrid Cloud

Data Provenance in Scientific Workflows

2011 ◽  
pp. 46-59
Author(s):  
Khalid Belhajjame ◽  
Paolo Missier ◽  
Carole Goble
Keyword(s):  
Real World ◽  
Scientific Workflow ◽  
Scientific Workflows ◽  
Data Provenance ◽  
Workflow Systems ◽  
Scientific Experiments

Data provenance is key to understanding and interpreting the results of scientific experiments. This chapter introduces and characterises data provenance in scientific workflows using illustrative examples taken from real-world workflows. The characterisation takes the form of a taxonomy that is used for comparing and analysing provenance capabilities supplied by existing scientific workflow systems.

Sign in / Sign up

Export Citation Format

Share Document