This document proposes a load balancing model for public clouds using cloud partitioning. It divides a large public cloud into partitions based on geographic location. When a job arrives, a main controller assigns it to the least loaded partition. Each partition uses algorithms like weighted round robin to further distribute jobs to nodes based on their calculated load degrees. The model aims to improve resource utilization and response times across the large, complex public cloud infrastructure.

1. LOAD BALANCING IN PUBLIC CLOUD USING
CLOUD PARTITIONING
Submitted by: under
guidance of:
Krishna Kumar (BE/1437/2011) Dr.
Amritanjali 1

2. TABLE OF CONTENTS:
1. Introduction to cloud computing
• What is cloud computing?
• Deployment models.
• Service models
• Advantages and Disadvantages
2. Load balancing in public cloud
• Architecture
• Models
• Algorithms
• Results
• Conclusion
2

3. INTRODUCTION to cloud computing
3

4. WHAT IS CLOUD COMPUTING?
Cloud:
Cloud refers to a Network or Internet. Cloud can provide services over
network.
Cloud Computing:
Cloud computing means "a type of Internet-based computing," where
different services — such as servers, storage and applications — are
delivered to an clients computers or devices through the Internet.
4

5. HOW CLOUD LOOKS:
5

6. SERVICE MODELS:
INFRASTRUCTURE AS A SERVICE (IAAS):
IaaS provides access to fundamental resources such as physical
machines, virtual machines, virtual storage, etc.
PLATFORM AS A SERVICE (PAAS):
PaaS provides the runtime environment for applications, development &
deployment tools, etc.
SOFTWARE AS A SERVICE (SAAS):
SaaS model allows to use software applications as a service to end users.
6

7. DEPLOYMENT MODELS:
Public Cloud:
The Public Cloud allows systems and services to be easily accessible to the general public. Public
cloud may be less secure because of its openness, e.g., e-mail.
Private Cloud:
The Private Cloud allows systems and services to be accessible within an organization. It offers
increased security because of its private nature.
Community Cloud:
The Community Cloud allows systems and services to be accessible by group of organizations.
Hybrid Cloud:
The Hybrid Cloud is mixture of public and private cloud. However, the critical activities are
performed using private cloud while the non-critical activities are performed using public cloud.
7

8. ADVANTAGES:
• Cost Efficient: All the maintenance works are done by cloud service
provider.Additional costs are minimised
• Mobility: Cloud mobility enables users to access anywhere with a
internet connection.
• Backup and Disaster Recovery: Most cloud providers offer backup and
recovery capabilities.
• Storage and Scalability: With the cloud,we have access to unlimited
storage and scalability.

9. DISADVANTAGES:
• Security and Privacy: Security is a major concern in clouds.Managing
confidential data may require deploying a private cloud.
• Compatibility: Softwares developed in one cloud may not be compatible
in other clouds.

10. Load balancing in public cloud
10

11. LOAD BALANCING:
• Load Balancing is a process of reassigning the total load to the
individual nodes of the collective system.
• It makes resource utilization effective and to improve the
response time of the job.
11

12. 12
• A large public cloud includes many nodes and the nodes
in different geographical locations.
• Cloud partitioning is used to manage this large cloud. A
cloud partition is a subarea of the public cloud with
divisions based on the geographic locations.
• The load balancing strategy is based on the cloud
partitioning concept.
SYSTEM MODEL:

13. AREA-WISE PARTITIONING OF CLOUD:
13

14. ARCHITECTURE:
14

15. 15
MAIN CONTROLLER AND BALANCER:
After creating the cloud partitions, the load balancing then starts:
• when a job arrives at the system, the main controller decides which
cloud partition should receive the job.
• The partition load balancer then decides how to assign the jobs to
the nodes.
• When the load status of a cloud partition is normal, this partitioning
can service the request.
• If the cloud partition load status is not normal, this job should be
transferred to another partition.

16. JOB ASSIGNMENT STRATEGY:
16

17. LOAD DEGREE CALCULATION:
17
• Load degree of node is related to various static and dynamic parameters
Static parameters:
• No of CPUs.
• CPU processing speed.
• Memory speed.
Dynamic parameters:
• Memory utilisation.
• CPU utilization.

18. 18
LOAD DEGREE OF NODES:
• Step 1: Define a load parameter set
• F = { F1 , F2 ,F3…, Fm}
where Fi’s are static & dynamic load parameters and m is total no of
parameters
• Step 2: Compute the load degree as
Load Degree (N) = ∑

19. 19
• Step 3: Three node load status levels are then defined as
• Idle When
Load Degree( N ) = 0
there is no job being processed by this node
• Normal When
0 < Load Degree(N)<= Load Degree
The node is normal and it can process other jobs.
• Overload When
Load Degree <= Load Degree( N)
The node is not available and can’t receive jobs until it returns to the normal.

20. 20
LOAD DEGREE OF PARTITION:
• Load degree of partition can be calculated as:
Load_degree avg =

21. 21
BEST PARTITION SEARCH ALGORITHM:
begin
while job do
SearchBestPartition(job);
if partitionState == idle || partitionState==normal then
send job to partition
else
search for another partition.
end if
end while
end

22. WEIGHTED ROUND ROBIN ALGORITHM:
22
STEP 1: Calculate the load degree of each node.
STEP 2: Make a circular queue of nodes ,arranged in ascending order
of their load degrees and queue ptr pointing to nodes with lowest
load degree.
Algorithm:
begin
while job do
Assign job to the node to which queue ptr is currently pointing to.
Move the queue ptr to next.
Re-calculate the load degree of the node and arrange nodes in circular queue.
end while
end

23. RESULT

24. 24

25. 25

33. CONCLUSION:
• Load balancing in cloud computing environment has an important
impact on the performance.
• In this project we have proposed a better load balance model for very
large and complex public cloud.
• With cloud partitioning concept it is possible to provide good load
balancing and hence improving the overall performance.

34. REFERENCES:
1. Microsoft Academic Research,Cloud computing,
http://libra.msra.cn/Keyword/6051/cloud-
computing?query=cloud%20computing,2012.
2. A load balancing model based on cloud partitioning for the public cloud
by Gaochao Xu, Junjie Pang and Xiaodong Fu.
3. http://www.javatpoint.com/cloud-computing-tutorial.
4. http://www.tutorialpoint.com/cloud_computing/
5. B.Adler,Load balancing in the cloud: Tools,tips and
techniques,http://www.rightscal.com/info_center/white-papers/Load-
Balancing-in-the-Cloud.pdf,2012.