See discussions, stats, and author profiles for this publication at: https://www.researchgate.

net/publication/260800636

Cloud Computing Simulation Using CloudSim

Article  in  International Journal of Engineering Trends and Technology · February 2014

DOI: 10.14445/22315381/IJETT-V8P216 · Source: arXiv

CITATIONS READS
37 429

2 authors:

Ranjan Kumar Gadadhar Sahoo

IIT Kharagpur Birla Institute of Technology, Mesra
10 PUBLICATIONS   146 CITATIONS    171 PUBLICATIONS   1,712 CITATIONS   

SEE PROFILE SEE PROFILE

Some of the authors of this publication are also working on these related projects:

Mirror Model View project

Correlate the microstructre with mechanical property by using EBSD Technique View project

All content following this page was uploaded by Ranjan Kumar on 13 April 2020.

The user has requested enhancement of the downloaded file.

 International Journal of Engineering Trends and Technology (IJETT) – Volume 8 Number 2- Feb 2014

Cloud Computing Simulation Using CloudSim

Ranjan Kumar#1, G.Sahoo*2

#
Assistant Professor, Computer Science & Engineering, Ranchi University, India
Professor & Head, Information Technology, Birla Institute of Technology, Mesra, India

Abstract— : As we know that Cloud Computing is a new deploying on real clouds. For simulation we need a
paradigm in IT. It has many advantages and disadvantages. But special toolkit named CloudSim. It is basically a
in future it will spread in the whole world. Many researches are
going on for securing the cloud services. Simulation is the act of Library for Simulation of Cloud Computing
imitating or pretending. It is a situation in which a particular set Scenarios. It has some features such as it support
of condition is created artificially in order to study that could for modeling and simulation of large scale Cloud
exit in reality. We need only a simple Operating System with
some memory to startup our Computer. All our resources will be Computing infrastructure, including data centers on
available in the cloud. a single physical computing node. It provides basic
classes for describing data centers, virtual machines,
Keywords— Cloud Computing, Data Center, VM Scheduler,
Host. applications, users, computational resources, and
policies.
I. INTRODUCTION
CloudSim supports VM Scheduling at two levels :
Cloud Computing infrastructure is different from First, at the host level where it is possible to specify
Grid Computing. It is the massive deployment of how much of the overall processing power of each
Virtualization technologies and tools. Hence, as core in a host will be assigned at each VM. And the
compared to Grid, Cloud has an extra layer as second , at the VM level, where the VMs assign
Virtualization, that acts as an execution and hosting specific amount of the available processing power
environment for cloud-based application services. to the individual task units that are hosted within its
To secure data in cloud is really very tough job. To execution engine.
understand the cloud computing we need to first
understand how these resources are placed in the Typical requirements and models of Cloud
cloud. Cloud Computing has basically two parts, Computing are Platform (PaaS), Software (SaaS),
the First part is of Client Side and the second part is Infrastructure (IaaS), Service-based application
of Server Side. The Client Side requests to the programming interface (API).
Servers and the Server responds to the Clients. The
request from the client firstly goes to the Master It has some Characteristics such as Low Cost
Processor of the Server Side. The Master Processor Software, Virtualization Service Orientation,
have many Slave Processors, the master processor Advance Security, Massive Scale, Resilient
sends that request to any one of the Slave Processor Computing, Geographic Distribution, Homogeneity,
which is free at that time. All Processors are busy in Broad Network Access, Rapid Elasticity, Resource
their assigned job and non of the Processor get Idle. Pooling etc.
Simulation opens the possibility to evaluate the
hypothesis prior to actual software development in There are basically four deployment models, they
an environment where one can reproduce tests. are Private Cloud, Public Cloud, Community Cloud,
Why we need simulation because it provides and Hybrid Cloud.
repeatable and controllable environment to test the
services. It tune the system bottlenecks before

ISSN: 2231-5381 http://www.ijettjournal.org Page 82

 International Journal of Engineering Trends and Technology (IJETT) – Volume 8 Number 2- Feb 2014

Cloud Infrastructure : In figure 2, we saw the life cycle of CloudSim. In

this, there are different stages from where
simulation have to pass each of the stage.

CloudSim Life Cycle :

Figure : 1

In figure 1, we saw infrastructure of Cloud, in

which Data Center consist of different Hosts and
the Host manages the VM Scheduler and VMs.

Cloudlet Scheduler determines how the available

CPU resources of virtual machine are divided
among Cloudlets. There are two types of policies
are offered :

1] Space-Shared (Cloudlet Scheduler Space

Shared) : To assign specific CPU cores to specific Figure : 2
VMs.
Organization of this paper is as follows: Related
2] Time-Shared (Cloudlet Scheduler Time Shared) : work is discussed in section II. Experimental setup
To dynamically distribute the capacity of a core and Result is discussed in section III. And section
among VMs. IV gives conclusion.

VmSchedular determines how many processing

II. RELATED WORK
cores of a host are allocated to virtual machines and
how many processing cores will be delegated to David S. Linthicum [1] described about the basic
each VM. It also determine how much of the information about the Cloud Computing and its
processing core’s capacity will effectively be various services and models like SaaS, IaaS, PaaS.
attributed for a given VM. He also described about the deploying models of
Cloud Computing and Virtualization services.

ISSN: 2231-5381 http://www.ijettjournal.org Page 83

 International Journal of Engineering Trends and Technology (IJETT) – Volume 8 Number 2- Feb 2014

Michael Miller [2] described about the various web Initialize the CloudSim Package :
based application related to Cloud Computing.
R.Bajaj and D.P. Agrawal [3] described about the Int num_user = 1; //number of cloud users
Scheduling. They said Scheduling is a process of
finding the efficient mapping of tasks to the suitable Calendar calendar = Calender.getInstance();
resources so that execution can be completed such
as minimization of execution time as specified by Boolean trace_flag = false; //mean trace events
customers. They described various types of
Scheduling like Static, Dynamic, Centralized, Initialize the CloudSim Library :
Hierarchical, Distributed, Cooperative, Non-
Cooperative Scheduling. They also described CloudSim.init(num_user, calendar, trace_flag);
Scheduling problem in Cloud and the types of users
like CCU ( Cloud Computing Customers) and Create Datacenters : These are the resource
CCSP ( Cloud Computing Service Providers). R.N. providers in CloudSim. We need at least one of
Calheiros, Rajiv Ranjan, Anton Beloglazov, C.A.F. them to run a CloudSim simulation.
De Rose, Rajkumar Buyya [4] described about the
Simulation techniques and the CloudSim. They Datacenter datacenter0 =
described the various features of CloudSim like it createDatacenter(“Datacenter_0”);
supports for modelling and simulation for large
scale of cloud computing infrastructure including Create Broker :
data centers on a single physical computing node. J.
Li, M. Qiu, X. Qin [5] described about the DatacenterBroker broker = createBroker();
optimization criterion that is used when making
scheduling decision and represents the goals of the Int brokerId = broker.getId();
scheduling process. The criterion is expressed by
the value of objective function which allows us to Create one Virtual machine :
measure the quality of computed solution and
compare it with different solution. C.H.Hsu and T. Vmlist = new ArrayList<Vm>();
L. Chen [6] described about the Quality of Service,
that is the ability to provide different jobs and users, VM description :
or to guarantee a certain level of performance to a
job. If the QoS mechanism is supported it allows Int vmid = 0;
the user to specify desired performance for their
jobs. In system with limited resources the QoS Int mips= 1000;
support results in additional cost which is related to
the complexity of QoS requests and the efficiency Long size = 10000; //image size (MB)
of the scheduler when dealing with them.
Int ram = 512; //vm memory (MB)

Long bw = 1000;
III. EXPERIMENTAL SETUP AND RESULT
To evaluate the performance of Cloud, results were Int pesNumber = 1; //number of cpu
simulated in Window 7 basic (64-bit), i3 Processor,
370 M Processor, 2.40 GHz of speed with memory String vmm = “Xen”; // VMM name
of 3 GB and the language used is Java. First code of
Simulation is tested on one Data Center with one Create VM :
Host and run on one Cloudlet.

ISSN: 2231-5381 http://www.ijettjournal.org Page 84

 International Journal of Engineering Trends and Technology (IJETT) – Volume 8 Number 2- Feb 2014

Vm vm = new Vm(vmid, brokerId, mips, pesNumber, Start Simulation :

ram, bw, size, vmm, new
CloudletSchedulerTimeShared()); CloudSim.startSimulation();

Add the VM to the vmList : CloudSim.stopSimulation();

vmList.add(vm1); Final step : Print result when simulation is over

Submit vm list to the broker : List<Cloudlet> newList =

broker.getCloudletReceivedList();
Broker.submitVmList(vmList);
printCloudletList(newList);
Creation of Cloudlets :
// Print the debt of each user to each datacenter
cloudLetList = new ArrayList<Cloudlet>();
Datacenter0.printDebts();
Cloudlet Properties :
The output of this simulation is :
Int id = 0; ========== OUTPUT ==========
Cloudlet ID STATUS Data center ID VM ID
PesNumber = 1; Time Start Time Finish Time
0 SUCCESS 2 0 400 0
long length = 400000; 400
*****PowerDatacenter: Datacenter_0*****
long fileSize = 300; User id Debt
3 35.6
long outputSize = 300; **********************************
After increasing the datacenter with two hosts and
UtilizationModel utilizationModel = new run two cloudlets on it. The cloudlets run in VMs
UtilizationModelFull(); with different MIPS requirements. The cloudlets
will take different time to complete the execution
Cloudlet cloudlet = new Cloudlet(id, length, depending on the requested VM performance.
pesNumber, fileSize, outputSize, utilizationModel, It prints the following output :
utilizationModel1, utilizationModel1);
========== OUTPUT ==========
Cloudlet.setUserId(brokerId); Cloudlet ID STATUS Data center ID VM ID
Time Start Time Finish Time
Cloudlet.setVmId(vmid); 1 SUCCESS 2 1 80 0
80
Add the Cloudlets to the list : 0 SUCCESS 2 0 160 0
160
cloudLetList.add(cloudlet1); *****PowerDatacenter: Datacenter_0*****
User id Debt
Submit cloudlet list to the broker : 3 224.8
**********************************
Broker.submitCloudletList(cloudLetList);

ISSN: 2231-5381 http://www.ijettjournal.org Page 85

 International Journal of Engineering Trends and Technology (IJETT) – Volume 8 Number 2- Feb 2014

IV. CONCLUSIONS REFERENCES

In this paper, we have proposed the codes for [1] David S. Linthicum “Cloud Computing and SOA Convergence in your
Enterprise”, 2011.
simulation. We see the various outputs in which [2] Michael Miller, “ Cloud Computing Web Based Application” 2012.
the information about the Cloudlets, Status, [3] R. Bajaj and D.P. Agrawal, “Improving Scheduling of Tasks in a
Heterogeneous Environment, IEEE Transaction on parallel and
Datacenter ID, Virtual Machine ID, Start Time, Distributed Systems”, p. 107-118, 2004.
Finish Time is given. And by changing the number [4] R. N. Calheiros, Rajiv Ranjan, Anton Beloglazov, C.A.F. De Rose,
Rajkumar Buyya, “CloudSim: A toolkit for modeling and simulation of
of Host, Datacenters and Cloudlets, we saw the cloud computing environments and evaluation of resource provisioning
difference. algorithms”, Software Practice and Experience, Wiley publishers, 2010.
[5] J.Li, M.Qiu, X.Qin, “Feedback Dynamic Algorithms for Preemptable
Job Scheduling in Cloud System”, IEEE, 2010.
[6] C.H.Hsu, T.L.Chen, “Adaptive Scheduling based on QoS in
Heterogeneous Environment”, IEEE, 2010.

ISSN: 2231-5381 http://www.ijettjournal.org Page 86

View publication stats