International Journal of Application or Innovation in Engineering & Management (IJAIEM)

Web Site: www.ijaiem.org Email: editor@ijaiem.org

Volume 8, Issue 12, December 2019 ISSN 2319 - 4847

Review on Workflow Scheduling Algorithm in

Cloud Platform
Prapti S.Deshmukh1, Prof.Sneha U.Bohra2
1
M.E scholar, Department of Computer Science and Engineering, G H Raisoni University,Amravati
2
Guide, Department of Computer Science and Engineering, G H Raisoni University,Amravati

Abstract
Cloud computing is the latest distributed computing paradigm and it offers tremendous opportunities to
resolve giant scale scientific issues. However, it presents various challenges that require to be self-addressed so as to be with
efficiency used for progress applications. The progress planning downside has been widely studied, there are very few initiatives
tailored for Cloud environments. A Cloud atmosphere differs from different distributed atmosphere in two major ways: on-
demand resource provisioning and pay-as-you-go rating model. The proposed system consists of a dynamic cost-effective
deadline-constrained heuristic algorithm for scheduling a scientific workflow in a public Cloud. This system aims to exploit the
advantages offered by the cloud computing for resource management in IT sector by implementing plan and schedule
algorithm to manage and allocate the resources by scheduling the deadlines of various tasks of major projects in IT industry
such as analysis , design ,implementation ,testing, deployment and maintenance .The proposed methodology also provides the
performance evaluation of an employee by analyzing his attendance, feedback from admin and their teammates based on his
work.
Keywords: distributed computing, heuristic , plan and schedule algorithm , performance evaluation

1. INTRODUCTION
Cloud is a general model for enabling on demand and a convenient way to access proof of computing resources which
includes servers ,networks, applications and services .The main features of cloud computing includes Data storage
,Data backup, Data accessibility, collaboration ,sharing network resources[1] .It provides various services over public
or private network . Applications such as email, web conferencing, customer relationship management , all run in
cloud. Cloud services are majorly categorized as: Infrastructure as a service (IaaS), which includes raw infrastructure
and associated middleware; Platform as a service (PaaS), which includes APIs for developing applications on an
abstract platform and Software as a service (SaaS) that provides support for remote software services. Clouds have a
broad impact on legacy scientific applications, because we anticipate that many existing legacy applications adapt to
and take advantage of new capabilities [6, 7]. Traditionally, developers of scientific applications have used local
workstations, super computers, clusters and grids platforms for running such workflows. Each of these platforms offer
various trade-offs in terms of usability, performance and cost. Cloud computing, has recently emerged as a promising
execution platform for huge and complex scientific applications [9]. PaaS based solutions of cloud involve the overhead
of porting legacy applications to new platforms whereas scientific computing SaaS services are currently rarely
available for usage. IaaS Cloud on the other hand, offers several cost and performance related benefits for executing
scientific applications as compared to traditional distributed execution environment.

Fig 1: Cloud Service delivery model

 Software as a Service(SaaS):-

Volume 8, Issue 12, December 2019 Page 34

International Journal of Application or Innovation in Engineering & Management (IJAIEM)
Web Site: www.ijaiem.org Email: editor@ijaiem.org
Volume 8, Issue 12, December 2019 ISSN 2319 - 4847

SaaS provides software application as a service for clients.. SaaS is some type of standard software package that can
be used on demand and is paid for by subscription, by use, by advertising, or by sharing information (such as email
addresses) of users. BigCommerce, Google Apps, Salesforce, Dropbox, MailChimp, ZenDesk, DocuSign, Slack,
Hubspot are some of the examples of SaaS.

 Platform as a Service(PaaS):-
PaaS offers runtime environment for many applications. Development & deployment tools, required to develop
application can be provided. AWS Elastic Beanstalk, Heroku, Windows Azure (mostly used as PaaS), Force.com,
OpenShift, Apache Stratos, Magento Commerce Cloud are some of the examples of PaaS.

 Infrastructure as a Service(IaaS):-
In IaaS cloud vendors deliver computational resources, storage and network as internet-based services. IaaS also
provides access to fundamental resources such as physical machines, virtual machines, virtual storage.

2. LITERATURE SURVEY:
Clouds are being explored as a solution to some of the problems with Grid computing, but the differences between
cloud computing and the Grid are often so diminished and obscured that they become indistinguishable from one
another. The term “Grid” computing was coined in the early 1990's to liken a distributed computing infrastructure to
the electrical power grid [1,2].Malawski et.al. proposed various dynamic and static algorithms for resource
provisioning and scheduling workflow ensembles in Cloud . These algorithms aim to maximize the number of executed
workflows while meeting the QoS constraints of deadline and budget. The proposed solutions acknowledge different
delays involved in leasing VM resources from the IaaS Cloud, such as VM acquisition and termination delays. Also,
the issue of performance variation of VMs is addressed by assuming that a task’s execution time may vary based on a
uniform distribution. They, however, consider only a single VM type ignoring the heterogeneous nature of IaaS Clouds
[3].Abusayeed ,Saifullah, David Ferry, Jing Li, Kunal Agrawal, Chenyang Lu, Christopher Gill studied about
Parallel Real time scheduling of DAG’s(directed acyclic graph) which proposed a novel task decomposition to
transform the nodes of a general DAG into sequential tasks. Since each node of the DAG becomes an individual
sequential task, these tasks can be scheduled either preemptively or non-preemptively [4]. Saeid Abrishami,
Mahmoud Naghibzadeh, and Dick H.J. Epema proposed a new QoS-based workflow scheduling algorithm based on
a novel concept called Partial Critical Paths (PCP),that tries to minimize the cost of workflow execution while meeting
a user-defined deadline. The PCP algorithm has two phases: in the deadline distribution phase it recursively assigns
subdeadlines to the tasks on the partial critical paths ending at previously assigned tasks, and in the planning phase it
assigns the cheapest service to each task while meeting its subdeadline[5].Chen and Zhang proposed an Ant Colony
Optimization algorithm with three QoS parameters: time, cost and reliability. Szabo at al.proposed a multi-objective
algorithm based on evolutionary approach, for execution of data-intensive scientific applications in a Cloud
environment such that the data transferred between tasks and the workflow execution time are minimized. Though,
these methods exhibit good performance they usually are more time consuming than heuristics approach[6].Mao and
Humphrey proposed a dynamic approach for scheduling workflow ensembles on Clouds such that all work-flows are
finished within their deadlines at minimum cost The algorithms presented by Mao and Humphrey were designed for
workflow ensembles and not for single workflow instances [7].Hoffa,Mehta ,and deore also studied the workflow
techniques in cloud computing. The virtual environment can provide good compute time performancebut it can suffer
from resource scheduling delays and wide area communications [8,9].Vinay K and S M Dilip Kumar also conducted
the experiment on the auto-scaling mechanism in cloud computing which is responsive to different Scientific workflow
and provide efficient resource utilization by considering idle-time of Virtual Machines [10,12].Sharif, Taheri and
Zomaya et.al proposed Online Multiple Workflow Scheduling under Privacy and Deadline in Hybrid Cloud
Environment by introducing OMPHC-PCPR as a further improvement in online multiple workflow scheduling
algorithms. It had evaluated the performance of the two algorithms and illustrated that OMPHC-PCPR outperforms
OPHC-TR by reducing the total cost by up to 50% [11].Mattess , Rodrigo N. Calheiros et.al proposed an application
scenario for the deadline-aware execution of MapReduce applications on hybrid Clouds. Although this approach lower
complexity, delivers good results. The connectivity between local and remote resources can be the solely dominant
factor affecting the completion time of the whole application if the Reduce tasks are not computationally intensive. The
proposed algorithm overcome all these issues [13].Lou and zhang designed a Dynamic Scheduling Strategy for Testing
Task in Cloud Computing which minimizes the makespan and realize the load balance. But , increased on cost of cloud
resource and the solution space in Genetic Algorithm[14].
3. EXISTING SYSTEM:

Volume 8, Issue 12, December 2019 Page 35

International Journal of Application or Innovation in Engineering & Management (IJAIEM)
Web Site: www.ijaiem.org Email: editor@ijaiem.org
Volume 8, Issue 12, December 2019 ISSN 2319 - 4847

The Existing Systems in Cloud platform shows the major issues such as:
 VM(virtual machine) performance variation.
 Resource acquisition delays
 Heterogeneous nature of cloud resources

4. CONCLUSION:
Cloud computing environment offers tremendous opportunities and alternatives to execute large scale industries
workflows Executing applications in Cloud involves making appropriate provisioning and scheduling decisions such
that the overall execution cost is minimized while meeting a user defined deadline. Towards this, a dynamic cost-
minimization and deadline constrained heuristic, plan and schedule algorithm, for scheduling of the major task in a
Cloud environment has been proposed in this work. In order to maintain low execution cost, resources are provisioned
just before they are needed. The objective of meeting the deadline is achieved through continuous monitoring of the
running tasks and dynamically making cost effective scheduling decisions for subsequent tasks such that the deadline
constraint is not violated. Extending the algorithm to include heuristics that ensure a task is assigned to a VM with
sufficient memory to execute it will be included in the algorithm. Finally, The Proposed system aim to implement
approach in a workflow engine so that it can be utilized for applications in real life environments.

References
[1] Jyoti Sahni and Deo Prakash Vidyarthi (2018) “A Cost-Effective Deadline-Constrained Dynamic Scheduling
Algorithm for Scientific Workflows in a Cloud Environment”, IEEE Transactions on Cloud Computing.
[2] Rodriguez M. A. & Buyya R, (2014) “Deadline based resource provisioning and scheduling algorithm for scientific
workflows on clouds”,IEEE Transactions on Cloud Computing.
[3] Malawski, M., Juve, G., Deelman, E., & Nabrzyski, J. (2012, November), “Cost-and deadline-constrained
provisioning for scientific workflow ensembles in iaas clouds.”,In Proceedings of the International Conference on
High Performance Computing, Networking, Storage and Analysis ,IEEE Computer Society Press.
[4] Saifullah, David Ferry, Jing Li, Kunal Agrawal, Chenyang Lu, Christopher Gill(2013), “Parallel Real-Time
Scheduling of DAGs”
[5] Abrishami, Naghibzadeh, Epema(2012 august). “Cost-Driven Scheduling of Grid Workflows Using Partial Critical
Paths”, IEEE TRANSACTIONS ON PARALLEL AND DISTRIBUTED SYSTEMS
[6] Wei-Neng Chen and Jun Zhang(JANUARY 2009). “An Ant Colony Optimization Approach to a Grid Workflow
Scheduling Problem with Various QoS Requirements”.
[7] Mao, M., & Humphrey, M. (2012, June). “A performance study on the vm startup time in the cloud. In Cloud
Computing (CLOUD), 2012 IEEE 5th International Conference on (pp. 423-430)”. IEEE.
[8] Deore, and Jaybhaye “A Survey on Workflow Scheduling in Cloud Computing”, IJIRSET.2017
[9] Hoffa ,Mehta, Timothy Freeman, Ewa Deelman, Kate Keahey, Bruce Berriman, John Good “On the Use of Cloud
Computing for Scientific Workflows” ,IEEE.
[10] Vinay K and S M Dilip Kumarz “Auto-scaling for Deadline Constrained Scientific Workflows in Cloud
Environment “(2016), IEEE.
[11] Sharif, Javid Taheri, Albert Y. Zomaya “Online Multiple Workflow Scheduling under Privacy and Deadline in
Hybrid Cloud Environment “2014 ,IEEE 6th International Conference on Cloud Computing Technology and
Science
[12] Chen, Li,Nanjing, Rub´en Ruiz’ “Cloud Workflow Scheduling with On-demand and Spot Block Instances”
,Proceedings of the 2017 IEEE 21st International Conference on Computer Supported Cooperative Work in
Design.
[13] Matess, Calheiros, and Rajkumar Buyya “Scaling MapReduce Applications across Hybrid Clouds” ( 2013) ,IEEE
27th International Conference on Advanced Information Networking and Application set Soft Deadlines’.
[14] Yang Lou, Tao Zhang, Jing Yan, Kun Li, Yechun” Dynamic Scheduling Strategy for Testing Task in Cloud
Computing” (2014) Sixth International Conference on Computational Intelligence and Communication Networks.

AUTHORS

Prapti S.Deshmukh currently pursuing M.tech (2nd year) from G h raisoni university, Amravati(Maharashtra) and
received the B.E degree from Sant Gadge baba Amravati University in 2016.

Volume 8, Issue 12, December 2019 Page 36

International Journal of Application or Innovation in Engineering & Management (IJAIEM)
Web Site: www.ijaiem.org Email: editor@ijaiem.org
Volume 8, Issue 12, December 2019 ISSN 2319 - 4847

Prof.Sneha U. Bohra is an Head of Department of Computer Science and Engineering in G h Raisoni

University,Amravati(Maharashtra).

Volume 8, Issue 12, December 2019 Page 37