You are on page 1of 20

Ant Colony Optimization for Load

Balancing in Cloud Computing

Guided by:

Submitted By:

Manas Ku Nanda
Asst Prof, Dept. of CA
ITER, SOAUniversity

Millan Prasad Mohanty


Branch: CSDP
Reg No:1461021006

03-Apr-16

Context
2

Introduction

Problem Statement/Objective

Motivation

Literature Survey

Proposed Model

Work Progress

Conclusion

References

03-May-16

Introduction
3

Cloud load balancing is the process of distributing workloads


across multiple computing resources in a cloud computing
environment.
To balance load distribution, we can migrate the load from the
heavy loaded nodes to the comparatively lightly loaded destination
nodes.
Goals of Load balancing:
To improve the performance substantially.
To have a backup plan in case the system fails even
partially.
To maintain the system stability
03-May-16

Ant Colony Optimization


4

The inspiring source of ACO is the foraging behavior of real ants.


When searching for food, ants initially explore the area surrounding
their nest in a random manner.

Indirect communication between the ants via pheromone trails


enables them to find shortest paths between their nest and food
sources. This characteristic of real ant colonies is exploited in ant
colony optimization.

03-May-16

Problem Statement/Objective
5

To design a Optimization algorithm using ACO technique to balance the


workload among different virtual machines and which Can check each and every virtual machines to determine whether
under loaded or overloaded
Can keep tract of the incoming tasks accordingly with virtual
machine's status.
Can Migrate any task from overloaded VM to under loaded VM.
Can able to increase overall performance of the system on the basis of
make span time.

03-May-16

Motivation
6

Motivations behind using this technique are:


ACO is Based on Positive Feedback approach, which accounts for
rapid discovery of good solutions.
Distributed computation avoids premature convergence.
The greedy heuristic helps find acceptable solution in the early
solution in the early stages of the search process.
Coupling ACO with local optimizers gave much more better solutions.
03-May-16

Literature Survey
7

Sl.
No
1

Authors Name /
Paper Title
Gupta Ekta
Deshpanda Vidya
A Technique Based
on Ant Colony
Optimization for
Load Balancing in
Cloud Data Center

Journal /
Conference
IEEE 13th
International
Conference
on
Information
Technology.

Year of
Publication
2014

Findings
Proposed a optimized
version of standard ACO
that coordinates load
balancer with optimizer.
Use of Cloud partitioning
helps to improve
performance of algorithm
as well as ease the
management of loads
among nodes.

03-May-16

Literature Survey (cont)


8

Sl.
No

Authors Name / Paper


Title

Kun Li,
Gaochao Xu,
Guangyu Zhao,
Yushuang Dong,
Dan Wang
Cloud Task
scheduling based on
Load Balancing Ant
Colony Optimization

Journal /
Conference
Sixth Annual
China Grid
Conference,
978-0769544724/11

Year of
Publication
2011

Findings
it works on minimization of
the task completion time that
are distributed among virtual
machine.
Comparison with the basic
ACO algorithm and FCFS
algorithm.
LBMM balanced the entire
system whether the entire task
are the same or not LBMM
can handle all type of
condition.

03-May-16

Literature Survey (cont)


9

Sl.
No
3

Authors Name / Paper


Title
R. Rastogi
Kumar Nishant,
Sharma Pratik
Load Balancing of
Nodes in Cloud Using
Ant Colony
Optimization

Journal /
Conference
14th International
Conference on
Computer
Modelling and
Simulation
(UKSim), IEEE,
pp: 3-8, March
2012

Year of
Publicatio
n
2012

Findings
Efficient distribution of
load such that ant never
dead.
It is centralized. So no
single point of failure.
It is best case that finds
the node is under loaded or
not.
The ant updated
the single result set rather
than updating their own
result set. It is compiled
only once duration of
movement of ant.
27-Mar-16
03-May-16

Literature Survey (cont)


10

Sl.
No
1

Authors Name /
Paper Title
S.C. Wang,
K.Q. Yan,
W.P. Liao,
S.S. Wang
Towards a Load
Balancing in a
Three-level Cloud
Computing
Network.

Journal /
Conference
3rd IEEE
International
Conference on
Computer
Science and
Information
Technology,
pp. 108-113

Year of
Publication
2010

Findings
Nodes are distributed in
three level structures
where work is distributed
among nodes.

It combine OLB
(Opportunistic Load
Balancing) to keep each node
busy & LBMM(Load
Balancing Min Min) to
achieve the minimum
execution time of each task

03-May-16

Literature Survey (cont)


11

Sl.
No
5

Authors Name /
Paper Title
Wei-Tao Wen
Chang-Dong Wang
De-Shen Wu
Ying-Yan Xie
An ACO-Based
Scheduling Strategy
on Load
Balancing in Cloud
Computing
Environment

Journal /
Conference

Year of
Publication

Ninth
International
Conference
on Frontier of
Computer
Science and
Technology

2015

Findings
the usage of pheromone is
maximized by establishing two
completely different traversing
strategy for ants. Violation of SLA
is minimized during migration.
The purpose is to let
the pheromone accumulate in the
PMs in high load condition
as much as possible, in order to
minimise live migrations when
new tasks come in queue.

03-May-16

Proposed Architecture
12

Our proposed module of load balancing consists 3 steps:

Find Overloaded node.

Select new Virtual machine for overloaded node

Place the virtual machine with node.

03-May-16

Proposed Architecture (cont)


13

1st step:

Input- N(no. of nodes in datacenter)


Output- Find Overload node
Begin
Initialize pheromone table
for Node 1 To Node N
pheromone(nodeid,threshold,cpuutilization)
enter nodeid
calculate cpuutilization
end
if cpuutilization > threshold
then
node is overloaded
select node id
end
end

03-May-16

Proposed Architecture (cont)


14

2nd Step:
Input- nodeid,m(no. of vm)
Output- select vm(vmid)
Begin
Select overload nodeid
for vm1 to m
then
calculate migrtation time
end
select vm=minimum migration time
end

03-May-16

Proposed Architecture (cont)


15

3rd Step:

Input virtual machine id(vmid),


underload node id(unid)
Output- Balanced the node

Begin
Select underload node from pheromone table(unid)
redirect the cloudlet of

vmid to the unid

if (node is balanced)
then
update pheromone table
end
en

03-May-16

Proposed Architecture (cont)


16

Fig 1:
overall system
flow chart

03-May-16

Work Progress
17

Programing language is JAVA.

Upto now we have setup the CoudSim environment with Eclipse IDE and
common math library has been integrated.

Creation od Data center, assigning properties, creation of Virtual


machines, hosts, cloudlets has been simulated using system given classes.

The inbuilt broker object is invoked which is responsible for assignment


of cloudlets into VMs.

The default migration policy is being examined.

Next work will be to implement our own migration policy by overriding


the default broker method.

03-May-16

Conclusion
18

This load balancing technique based on Ant Colony Optimization


gives optimal resource utilization.

The performance of the system is enhanced with high availability of


resources, thereby increasing the throughput.

The main benet of this approach lies in its successful detections of


overloaded and underloaded nodes and thereby performing
operations based on the identied nodes.

The advantage of the approach lies in the fact that the task of each
ant is specialized rather than being general and the task depends on
the type of rst node that was encountered whether it was
overloaded or under loaded.

03-May-16

References
19

[1] Ekta Gupta , Vidya Deshpande . A Technique Based on Ant Colony Optimization for
Load
Balancing in Cloud Data Center ,IEEE 13th International Conference on
Information Technology. 2014
[2] Rastogi Ravi, Kumar Nishant, Sharma Pratik, Vishal Krishna, Chhavi Gupta, Load
Balancing of Nodes in Cloud Using Ant Colony Optimization,IEEE14th International
Conference on Modelling and Simulation. 2012
[3] Klaithem Al Nuaimi, Nader Mohamed, Mariam Al Nuaimi and Jameela AlJaroodi .A Survey of Load Balancing in Cloud Computing: Challenges and Algorithms.
IEEE Second Symposium on Network Cloud Computing and Applications. 2012
[4] Kun Li, Gaochao Xu, Guangyu Zhao, YushuangDong, Dan Wang, Cloud Task
scheduling based onLoad Balancing Ant Colony Optimization, Sixth Annual ChinaGrid
Conference, 978-0-76954472-4/11. 2011
[5] S.C. Wang, K.Q. Yan, W.P. Liao and S.S. Wang, Towards a Load Balancing in a
Three-level Cloud Computing Network, Proceedings of the 3rd IEEE International
Conference on Computer Science and Information Technology, pp. 108-113, 2010

03-May-16

20

Thank You

03-May-16

You might also like