Download Load Balancing Performance in Content Delivery Networks (CDN`s)

ISSN : 0976-8491 (Online) | ISSN : 2229-4333 (Print)
IJCST Vol. 5, Issue 4, Oct - Dec 2014
Load Balancing Performance in Content
Delivery Networks (CDN’s)
1
Kanthai Srinivasa Rao, 2Manjula Srinivas, 3Molli Srinivasa Rao
1,2,3
Dept. of CSE, VITAM College of Engineering, Andhra Pradesh, India
Abstract
It is an interesting issue of describing and executing an effective
law for load balancing in Content Delivery Networks (CDNs).
A CDN system conceded out through the exploitation of a fluid
flow model classification of the network of servers. Beginning
from such categorization, we develop and demonstrate a lemma
about the network queues balance. The result is then leveraged
in mandate to create a novel distributed and time-continuous
algorithm for load balancing which is also reformulated in a timediscrete form. The isolated devising of the proposed balancing law
is ultimately deliberated in terms of its definite application in a realworld consequence. Lastly, the complete method is authenticated
by means of imitations.
Keywords
Content Delivery Network (CDN), Control Theory, Request
Balancing
I. Introduction
A CDN contains of a unique server called as back-end server
comprising new data to be dispersed together with one or more
dissemination servers called as surrogate servers. The surrogate
servers are dynamically efficient by the back-end server. Surrogate
servers are characteristically used to store static data, while dynamic
information (i.e., data that change in time) is just stored in a small
number of back-end servers. In some characteristic situations,
there is a server called redirector, which dynamically redirects
client requests based on designated policies. The furthermost
important performance enhancements derived from the adoption
of such a network concern two aspects. Firstly, overall system
amount that is the average number of requests served in a time unit
which is improved also on the basis of the processing capabilities
of the available servers. Secondly, response time proficient by
clients after issuing a request.
II. Related Work
Active load-balancing approaches characterise a valid substitute
to static algorithms. Such methods make use of information
approaching either from the network or from the servers in order
to advance the request assignment process. The selection of the
suitable server is done through a assortment and succeeding
analysis of various parameters extracted from the network
elements. Depending on how the scheduler interrelates with
the other mechanisms of the node, it is conceivable to classify
the balancing algorithms in three fundamental models a queueadjustment model, a rate-adjustment model and a hybridadjustment model.
III. Existing Method
In a rate-adjustment model, in its place the scheduler is situated
just before the local queue. As per the new request the scheduler
chooses whether to allocate it to the local queue or send it to a
remote server.
In a hybrid-adjustment approach for load balancing, the scheduler
is permitted to switch both the inward appeal rate at a node and
148
International Journal of Computer Science And Technology
the local queue length. Therefore in Existing systems, a new
request indeed a CDN server can both intricate locally the request
or forward it to other servers according to a definite decision
rule which is grounded on the state information swapped by the
servers. Such a method bounds state switching above to fair local
servers.
IV. Disadvantages
A precarious constituent of CDN design is the request routing
mechanism. It permits to direct user requests for content to the
appropriate server based on a specified set of parameters. The
routing process related with a request might take into account
various constraints like traffic load, bandwidth, and servers’
computational capabilities in order to deliver the best performance
in terms of time of service, delay, etc.
V. Proposed Method
An appropriate load-balancing law that promises equilibrium of
the queues in a balanced CDN by means of a fluid flow model
for the network of servers. We present a new mechanism for
readdressing inward customer requirements to the furthermost
suitable server, thus corresponding the overall system requests
load. The mechanism leverages local balancing in demand to
attain global balancing. This is carried out through a periodic
collaboration among the system nodes.
VI. Advantages
The outstanding performance of mechanism might be paid in
terms of an important number of redirections. Subsequently the
redirection process is communal to all the algorithms analyzed,
we completely appraise the proportion of requests redirected more
than once over the total number of requests generated.
VII. Content Delivery Network
Fig. 1:
VIII. Modules
A. Network Formation
The Content Delivery Network is formed in network formation
module. The network comprise back end server, redirect server,
number of surrogate server and number of client. Every client
links with one surrogate server. The Back end server uphold all
server as redirect and surrogate server.
w w w. i j c s t. c o m
ISSN : 0976-8491 (Online) | ISSN : 2229-4333 (Print)
B. Fluid Queue Model
Every server preserve an own queue for unloading the client
request. Let qi(t) be the queue occupancy of server i at time t. Study
the instant arrival rate αi(t) and the instant service rate δi(t).
The fluid model of CDN servers queues is given by
C. Load Balance
The client sends the request to the server. The server initially check
the queue, if queue limit exceeds it redirect the server to others,
else it respond the user request. Each T seconds, the server directs
its status information to its neighbors and at the same time waits
for their information. When a new request attains at a server, it
confirms the presence of neighbors with a lower load. If no such
neighbors are present, the server locally procedures the request
and serves it; otherwise the balancing strategy is implemented. For
requests rearrangement, we implement a random number generator
with identical distribution between 0 and 1. Conditional on which
interval the generated number falls in, the algorithm selects the
corresponding peer for redirecting the incoming request.
Algorithm Used
Distributed Load-Balancing Algorithm
w w w. i j c s t. c o m
IJCST Vol. 5, Issue 4, Oct - Dec 2014
X. Conclusion
Characterisation of an algorithm that objectives at achieving load
balancing in the network by eliminating local queue instability
conditions through redeployment of potential excess traffic to the
set of neighbours of the crowded server. The algorithm is first
presented in its time-continuous formulation and then put in a
discrete version precisely conceived for its actual execution and
deployment in an operational scenario. The present work signifies
for us a first step to the consciousness of a complete solution for
load balancing in a cooperative, distributed environment.
References
[1] S. Manfredi, F. Oliviero, S. P. Romano,“Distributed management for load balancing in content delivery networks”, in
Proc. IEEE GLOBECOM Workshop, Miami, FL, Dec. 2010,
pp. 579–583.
[2] H. Yin, X. Liu, G. Min, C. Lin,“Content delivery networks:
A Bridge between emerging applications and future IP
networks”, IEEE Netw., Vol. 24, No. 4, pp. 52–56, Jul.–Aug.
2010.
[3] J. D. Pineda, C. P. Salvador,“On using content delivery
networks to improve MOG performance”, Int. J. Adv. Media
Commun., Vol. 4, No. 2, pp. 182–201, Mar. 2010.
[4] D. D. Sorte, M. Femminella, A. Parisi, G. Reali,“Network
delivery of live events in a digital cinema scenario”, In Proc.
ONDM, Mar. 2008, pp. 1–6.
[5] Akamai,“Akamai”, 2011 [Online] Available: http://www.
akamai. com/index.html
[6] Limelight Networks,“Limelight Networks”, 2011 [Online]
Available: http://.uk.llnw.com
[7] CDNetworks,“CDNetworks”, 2011 [Online] Available:
http://www.us.cdnetworks.com/index.php
[8] Coral,“The Coral Content Distribution Network”, 2004
[Online] Available: http://www.coralcdn.org
[9] Network Systems Group,“Projects”, Princeton University,
Princeton, NJ, 2008. [Online] Available: http://nsg.
cs.princeton.edu/projects
[10]A. Barbir, B. Cain, R. Nair,“Known content network (CN)
re- quest-routing mechanisms”, IETF, RFC 3568 Internet
Draft, Jul. 2003, [Online] Available: http://tools.ietf.org/html/
rfc3568
[11] T. Brisco,“DNS support for load balancing”, IETF, RFC
1794 In- ternet Draft, Apr. 1995 [Online] Available: http://
www.faqs.org/rfcs/rfc1794.html
[12]M. Colajanni, P. S. Yu, D. M. Dias,“Analysis of task assignment
policies in scalable distributed Web-server systems”, IEEE
Trans. Parallel Distrib. Syst., Vol. 9, No. 6, pp. 585–600,
Jun. 1998.
[13]D. M. Dias, W. Kish, R. Mukherjee, R. Tewari,“A scalable
and highly available Web server”, In Proc. IEEE Comput.
Conf., Feb. 1996.
International Journal of Computer Science And Technology 149
ISSN : 0976-8491 (Online) | ISSN : 2229-4333 (Print)
IJCST Vol. 5, Issue 4, Oct - Dec 2014
Kanthai Srinivasa Rao is pursuing
his M.Tech in Computer Science and
Engineering from VITAM College
of Eng affiliated to JNTU Kakinada.
He received his B.Tech degree from
Avanthi Institute of Engineering &
Technology JNTU Hyderabad, Andhra
Pradesh, India. His research interests
include computer and network
security.
Manjula Srinivas. She received her
M.Tech in Computer Science and
Engineering from Avanthi Institute
of Engineering & Technology,
Vizianagaram in 2012. She received
her B.E from K.V.G. College of
Engineering,Sullia, Karnataka State.
She is currently working as Assistant
Professor, CSE Dept. in VITAM College
of Engg., Andhra Pradesh, India. Her
research interests include computer
network architectures, algorithms and
protocols, Sensor Networks,mobile Ad-hoc Networks,Wireless
Mesh Networks, computer and network security.
Molli Srinivasa Rao is pursuing his
Ph.D from Andhra University. He
received his M.Tech in Computer
Science and Technology from Andhra
University in 2003. He received
his B.Tech degree from C.B.I.T.,
Hyderabad in 2001. He is currently
working as Associate Professor, CSE
Dept. in VITAM College of Engg.,
Andhra Pradesh, India. His Research
Interests include mobile Ad-hoc
Networks, Sensor Networks,Wireless Mesh Networks, computer
and network security.
150
International Journal of Computer Science And Technology
w w w. i j c s t. c o m