LOAD BALANCING ON HETEROGENOUS

SYSTEMS USING
EVOLUTIONARY TECHNIQUES

SUBMITTED BY :-
AASTHA MEDIRATTA (0709110001)
ABHINAV BHATNAGAR (0709110002)
ABHISHEK KANAUJIA (0709110004)
KHUSHBU SRIVASTAVA (0709110050)
 MOTIVATION
Load balancing is a technique to distribute workload evenly
across two or more computers, network links, CPUs, hard drives,
or other resources, in order to get optimal resource utilization,
maximize throughput, minimize response time, and avoid overload.
The load balancing service is usually provided by a dedicated
program or hardware device.
It is commonly used to mediate internal communications
in computer clusters, especially high-availability clusters. If the
load is more on a server, then the secondary server takes some load
while the other is still processing requests.
The increasing popularity of the World Wide Web has resulted in
large bandwidth demands which translate into high latencies
(response time) perceived by Web users. An effective load
sharing and balancing mechanism becomes a necessity in such
cases.

It is critical to improve the performance of the web to make it as

reliable as the telephone and hence a suitable medium for high
volume business critical applications. This problem faced by the
web servers working in heterogeneous environment, of managing
the load optimally without leading to web failure is the prime
motivation for our project.
 OBJECTIVE
Load balancing is the process by which inbound internet protocol traffic
can be distributed across multiple servers. The Project Objectives
include:
 Designing optimization Algorithms essentially to equally spread the
balancing load on processors and maximize their utilization while
minimizing the total task execution time.

 To minimize the total execution and communication cost encountered

by the task assignment, subject to the resource constraints.

 To dynamically schedule the tasks in a heterogeneous environment.

 A scheduling strategy is to be presented which uses PSO (Particle

Swarm Optimization) to schedule heterogeneous tasks on to
heterogeneous processors to minimize the total execution time. It
operates dynamically, which allows the new tasks to arrive at any time
interval.
 SCOPE OF THE PROJECT

1. Understanding load balancing problem faced by web servers in

a heterogeneous environment.

2. Understanding various available optimization algorithms.

3. Implementation of Particle Swarm Optimization Algorithm and

Hybrid Particle Swarm Optimization Algorithm.

4. Generation of initial swarm.

5. Evaluation of Initial swarm using Fitness Function.

Evaluation of Fitness Function.
 TECHNICAL FEASIBILITY
We want to develop a system, where the internal details of
processing will be hidden from user and will be powerful
enough to deliver the required functionality.

To bring in the required functionality and speed in

development of the application, this system is developed using
MATLAB. The entire scope of project is valid on any windows
with network infrastructure. The project has not been extended
to execute on multiple platforms such as FreeBSD and
Windows of different servers. Since the network is
continuously changing, it is difficult to evaluate the
performance of different servers associated with the cluster.
 
 REFERENCES

[1] P Visalakshi , S N Sivanandam, “Dynamic Task Scheduling with Load Balancing using
Hybrid Particle Swarm Optimization”, Int. J. Open Problems Compt. Math., Vol. 2, No. 3,
September 2009,ISSN 1998-6262.
[2] Aly E. El-Abd, Mohamed I. El-Bendary, “A Neural Network Approach for Dynamic
Load Balancing In Homogeneous Distributed Systems”, IEEE Trans. Parallel and distributed
systems, Vol 9, No.6, June 1997.
[3] Albert Y. Zomaya, Senior Member, IEEE, and Yee-Hwei The, “Observations on Using
Genetic Algorithms for Dynamic Load-Balancing”, IEEE Transactions On Parallel And
Distributed Systems, Vol. 12, No. 9, September 2001.
[4] Swagatam Das, Ajith Abraham, and Amit Konar, “Particle Swarm Optimization and
Differential Evolution Algorithms: Technical Analysis, Applications and Hybridization
Perspectives”.
[5] Chunming Yang, Simon D, “A new particle swarm optimization technique”,
Proceedings of the International Conference on Systems Engineering (2005),pp.164-169.
[6] Parsopoulos K.E, and Vrahatis M.N, “Recent approaches to global optimization
problems through particle swarm optimization”, Natural Computing , Vol.1(2002),
pp. 235 – 306.