Scheduling Algorithm Based Simulator for Resource Allocation Task in Cloud Computing

International Research Journal of Engineering and Technology (IRJET) e-ISSN: 2395-0056
Volume: 04 Issue: 07 | July -2017 www.irjet.net p-ISSN: 2395-0072
© 2017, IRJET | Impact Factor value: 5.181 | ISO 9001:2008 Certified Journal | Page 3129
Scheduling Algorithm Based Simulator for Resource Allocation
Task in Cloud Computing
H. Rifaya Baswan1, Dr. A. Nagarajan2
1M.phil scholar, department of computer applications, Alagappa university, Karaikudi, Tamilnadu, India.
2Assistant professor, department of computer applications, Alagappa university, Karaikudi, Tamilnadu, India.
---------------------------------------------------------------------***---------------------------------------------------------------------
Abstract - In emerging trends in cloud computing to build
and maintain the cloud infrastructure with flexible by
virtualization technology. Virtualization context, it offers the
resource allocation, virtualization flexible platform, and
reliability. It provides the resource allocation to requesting
user needs flexibly. In this resource allocation, considering
satisfy the most of the requesting users. It is the resources
based on scheduling process done by service level agreement.
In this process priority based allocation, it contains allocating
resource for priority based. Its occurs the starvation to the
lowest priority process it is leave the low priority task. We
propose the allocating resource based PERT technology
scheduling algorithm and multiple SLA parameters such as to
allocate memory space, it locating distributed load to the
equal nodes. The capacity of physical machine to satisfy the
all needs of all virtual machine. Inthisexperimentalresults the
process of where the resources obtained in our proposed
algorithm to provide efficiency in scheduling allocating
process for all tasks.
Key Words: virtualization technique, PERT based
scheduling algorithm, resource allocation.
1. INTRODUCTION
In this cloud computing in advanced information and
technology, computing Paradigm largely connected to the
public and private network, it is dynamically provided
information to be resources and storage files. With this
technology in this evaluate cost, computation cost, hosting
application, content storage and performance of delivery
time is reduced. Cloud computingexperimentallyproved the
cost and efficient, where the data shared by software and
information send over the network. Cloud computing
provided information via the internet, which are getting
from web browser, while the performance software and
information are stored on servers at remote location.
Scheduling process to be highly efficient manner and
proper virtual machine as per the SLA performance for each
process and at the same time performance is high, resource
management is the important concept in the cloud
computing it consists of factors are cost management,
performance, efficiency they are affected by resource
management. Resource management mainly concentrates
with allocating resource and scheduling task.
When a job requesting to the cloud, it is performed
different tasks. In parallel processing in this task 1) the job
requesting how the resources to be allocating 2)cloudwhat
process to execute in cloud. Resource management model it
is executed task to be interconnected with shared resources
and task in this process to be workflow application model it
represented by the directed acyclic graph (DAG) it is the
process of nodes and edges denoted from the task, in this
paper we propose dynamic resource allocation to overcome
the workload in application using PERT technology to plan,
schedule and large task to control. The process of lower
priority and higher priority advanced task to perform
workload from the waiting queue.
In this PERT know about the input details from thecloudbut
cannot know certainly the cloud how to allocate resource in
cloud. DAG this approach uses the relation to compare
network it identify the queries related to the cloud
resources. Priority scheduling process to allocate the
completion of the task and decide by total completion of the
task and earliest completion of the task. In this PERT
algorithm, higher priority from the critical and non-critical
task it will be executed. And comparison of the priority and
our proposed to equalize the lower priority to take time
completion.
2. RELATED WORK
Resource allocation, resource scheduling andjobscheduling
algorithms are proposed in the cloud computing. [1] In
proposed performance of the optimization model in
workflow application, it is presented and performance is
better for completion time and allocating resource. [2]
Scheduling and scheduler to get resources in cloud, optimal
cloud computing in preceding the allocating do the job best
machines and minimize the lower priority and execution
cost. [3] In this allocating resources and scheduling
workflow process in data center, it also considers execution
time and cost expenditure are both. In this paper model,
infrastructure as a service architecture to be performed and
presentation of the priority based algorithm it to be
allocated resources to be priority wise. Itisprocessofhigher
priority to finish first and its starvation of the lower priority
that is the priority to satisfy them. [4] Proposed algorithm
and then efficient memory allocation using MPS (memory
processor storage) it is a matrix model. Priority based
algorithm it is a minimum wastage and maximum profit. [5]

Workflow scheduling algorithmisbasedon qualityofservice
and user defined to know the critical path method and
minimize the cost in workflow execution. And a cost based
resource allocation scheduling algorithm it is used the
market theory to resource schedule to refer user’s
requirement. [6] In a priority based scheduling algorithm it
is performed by the dynamic allocation of the jobs to be
processed multiple SLA in cloud jobs.
3. PROPOSED SCHEDULING ALGORITHM
The scheduling modelling algorithmisdirectedacyclicgraph
(DAG) G (T, E), it T represents the set of n tasks and E
represents the set of all directed edges m. A DAG it is calleda
task graph, it nodes represent a task of the workflow
application and edges represent the edges the relationship
between these tasks. Each edge is tij=(ti,tj) ∈E it between
these tasks.it ti and tj represents inter tasks communication
and precedence. And task ti must complete execution before
task tj.
In this task graph, it must completed from the before task
and follows successor of the completion of the given task.
Each successor of a task if it is completion of the giventask,a
task graph without parent it is called an entry task and task
without child it is called an exit task. This is an algorithm for
a single entry task and a single exit task, and two dummy
tasks tentry and texit is added from the beginningand endof
the task graph. These are the dummy task have a zero
execution time and they connected with the zero weighted
edges to the entry and exit tasks.
Sequentially connected to the task nodes in the task graph
from the entry node to exit task, is called path. The path of
the length is measured by the sum of weights on the task
edges on the path graph.
The longest path is called critical path and nodes of the
corresponding task is called critical task nodes andthatthey
must be completed as scheduled to meet by the scheduled
time. If the entire process is completed it is a ready task.
Transmission of data from the task is represented by the
communication time CT (ti,tj) is the time and it is
represented by the matrix form CT [ti,tj]n*n, n represents
the number of tasks the edge of the weight, wij it represent
the communication time CT (ti,tj) between two tasks of ti
and tj.
A logical unit of work of a task ti is executed by a resource.
The estimated execution time tocompletetask tibyresource
rj gives from the execution time can be represented by a
form of matrix form, ET [ti,rj]n*m itnrepresentsthenumber
of task and m represents the number of resources. The task
is to wait for until allocated resource completes the
execution of the current task. In case, waiting time, WT [ti,tj]
of task ti, and resource rj is zero, if the resource is process
means does not executing another task.
In this paper, represented by problem of the task duration
where low priority leave the process and it the waiting time
is long, it cannot be the execution certainly and that the
schedulingprocessthroughnormal probabilitydistributions.
4. SCHEDULING PROCESS
4.1 Problem Definition
In cloud computing for priority based scheduling algorithm
for resources allocates from cloud computing resources in
effective, efficient, and optimized task. This algorithm
contains that a set of resources is certainly distributed and
heterogeneous in nature. Priority assign by the task of
completion of time and earliest finished time of task.
4.2 Problems Identifying the Scheduling Task
In this approach, tasks duration is not known certainly. The
task duration is performed to have normal distribution, this
is the task is distribution task duration.
EX: randn(1,20000) this means two thousand number in
random on normal distribution
4.3 Directed Acyclic Graph Scheduling Algorithm
1) First give the input n task and m resources
2) N task in the DAG
3) Its communication is randomly
4) ET [ti,rj]n*m.
5) Its resources in randomly available time.
6) Its task i=1 to n
7) Find the execution process is minimum priority for
all task is optimized.
5. RESULTS
In this paper, priority based scheduling model its
performance is evaluated. A comparison of the algorithm
first is priority based algorithm, second is PERT based
scheduling technique, and third is priority and PERT based
scheduling algorithm. Mainly priority based algorithm to be
executed task and then secondlyPERTmodel firstlyexecutes
all critical tasks it have a lowest completion time and then
and then non critical tasks with lower completion time and
then execute the higher priority to the critical task and then
execute non critical task to the higher priority and
considering the various tasks in completion time.
5.1 Comparative analysis
Analyses Task 1
Firstly performed by the small workflow network for DAG,
below the figure as shown in 1: that have an eight numbers

of task (nodes) and nine numbers of activities among these
task.
Fig -1: Example for small workflow for DAG
Analyses Task 2
Secondly performed by the large workflownetwork forDAG,
below the figure as shown in 1: that have an eleven numbers
of task (nodes) and twenty numbers of activities among
these task.
Fig -2: Example for large workflow for DAG
Analyses Task 3
Third analysis is performed by the number of tasks
performed by the workflow network for DAG, below the
figure as shown in 3 and 4: that have a large workflow has a
small makespan and small completion time as compared to
small workflow network.
Chart -1: Makespan
Chart -2: Completion Time
6. CONCLUSIONS
Resource management technique it is one of the important
issues to be solved in cloud computing.Inthispaper,priority
based algorithm is the proposed model. It is compared with
other scheduling model for lower priority task to optimize
the completion time. Results of the proposed algorithm it is
efficiency and effectiveness of the proposed model.Itreduce
the completion time, in future work critically in priority
model will be to make more effective and efficiencymodel in
resource management in cloud computing.
ACKNOWLEDGEMENT
I would like to thank Dr.A.Nagarajan for his guidance and
support for preparing this paper.

REFERENCES
[1] M. Guzek, P. Bouvry, E.-G. Talbi, "A survey of
evolutionary computation for resource management of
processing in cloud computing [review article]",
Computational Intelligence Magazine IEEE, vol.10,no.2,
pp. 53-67, 2015.S.
[2] S. Manvi, G. K. Shyam, "Resource management for
Infrastructure as a Service (IaaS) in cloud computing: A
survey", Journal of Network and Computer Applications,
vol. 41, pp. 424-440, 2014.
[3] S. K. Garg, R. Buyya, and H. J. Siegel, “Time and cost trade
off management for scheduling parallel applications on
utility grids,” Future Generation. Computer System,
26(8):1344–1355, 2010.
[4] M. Salehi and R. Buyya, “Adapting market-oriented
scheduling policies for cloud computing,” In Algorithms
and Architectures for Parallel Processing, volume 6081
of Lecture Notes in Computer Science, pages 351–362.
Springer Berlin / Heidelberg, 2010.
[5] J. M. Wilson, “An algorithm for the generalized
assignment problem with special ordered sets,” Journal
of Heuristics, 11(4):337–350, 2005.
[6] M. Qiu and E. Sha, “Cost minimization while satisfying
hard/soft timing constraints for heterogeneous
embedded systems,” ACM Transactions on Design
Automation of Electronic Systems (TODAES), vol. 14, no.
2, pp. 1–30, 2009.
[7] M. Qiu, M. Guo, M. Liu, C. J. Xue, and E. H.-M. S. L. T. Yang,
“Loop scheduling and bank type assignment for
heterogeneous multibank memory,” Journal of Parallel
and Distributed Computing(JPDC), vol. 69, no. 6, pp.
546–558, 2009.
[8] A. Dogan and F. Ozguner, “Matching and scheduling
algorithms for minimizing execution time and failure
probability of applications in Heterogeneous
computing,” IEEE Transactions on Parallel and
Distributed Systems, pp. 308–323, 2002.
[9] T. Hagras and J. Janecek, “A high performance, low
complexity algorithm for compile-time task scheduling
in heterogeneous systems,” Parallel Computing, vol. 31,
no. 7, pp. 653–670, 2005.
[10] J. M. Wilson, “An algorithm for the generalized
assignment problem with special ordered sets,” Journal
of Heuristics, 11(4):337–350, 2005.
BIOGRAPHIES
Rifaya Baswan .H is a M.Phil
Scholar from the department of
computer applications, Alagappa
University, Karaikudi, Tamil Nadu,
India.
1’st A
uthor
Photo

Scheduling Algorithm Based Simulator for Resource Allocation Task in Cloud Computing

Recommended

More Related Content

What's hot (20)

Similar to Scheduling Algorithm Based Simulator for Resource Allocation Task in Cloud Computing (20)

More from IRJET Journal (20)

Recently uploaded (20)

Scheduling Algorithm Based Simulator for Resource Allocation Task in Cloud Computing