Sie sind auf Seite 1von 4

International Journal of Emerging Trends & Technology in Computer Science (IJETTCS)

Web Site: www.ijettcs.org Email: editor@ijettcs.org, editorijettcs@gmail.com Volume 3, Issue 1, January February 2014 ISSN 2278-6856

Comparative analysis of proposed algorithm with existing load balancing scheduling algorithms in Cloud Computing
1

Amritpal Singh, 2 Sunil Kumar Gupta, 3 Rajeev Kumar Bedi


1

Assistant Professor, Global Institute of Management, Amritsar 2,3 Associate Professor, BCET, Gurdaspur

Abstract: Now a days, internet is the basic need in every


segment of life. Every sector is using different web applications on internet, with the help of which they are fulfilling their jobs. At the same time, number of customers using the same application, as a result traffic increases on servers. For availability and scalability, load balancing is an important concern as to enhance the performance of machines in the virtual environment. There are certain ways with the help of which performance can be enhanced. We can can improve the performance by providing higher configuration machines to the users or by introducing a new load balancing algorithm in cloud computing. In this paper, we deploy a new load balancing algorithm in cloud computing and compare the different parameters as CPU processing time, response time and resource allocation time between the existing load balancing algorithm.

Keywords:- Load Balancing, CPU processing time, Response Time.

1. INTRODUCTION
In todays world, with the increase in demand of applications, load on servers increases day by day. With the introduction of E-Commerce, this leads to many business activities carry out with the help of internet. As a result, with the use of particular web application, providers of web application want to ensure the availability of access of their application and guarantee that their requests are processed as soon as possible [7]. Cloud computing is also a new boom in IT sector. With the help of Cloud Computing, a user dont need to spend a lot of money on the infrastructure, a user can request as much infrastructure as it is required [8]. There is an important module in Cloud Computing, pay-per-use. It means a user will pay to the service provider as much services he/she is using. But the demand of Cloud Computing services also increases. Load Balancing is the process by which internet traffic can be distributed across number of servers. Traditional load balancing techniques are not much efficient as to provide healthy environment as per todays requirement. Most of the loads balancing techniques used these days are static [7], not suitable according to todays environment. By survey the pro and cons of different load balancing algorithms, dynamic load balancing algorithms can only manage the load on different servers [11]. Volume 3, Issue 1 January February 2014

1.2 Cloud Deployment Models There are four types cloud Deployment models [10] that widely used are: Private: This model is not much concerned with cost efficiency; it is related with buying, building and managing your own infrastructure. But it is considered as the safest model in terms of security. During their initial adaptations to the Cloud, data security is the main challenge to many organizations. So, now the main focus is on security of the data and other meaningful information of the customer. Examples: - SugarCRM and Amazon are two models that are used in Private Cloud. Public: In this model, services and infrastructure are provided to various clients. This service is provided free of charge or on the basis of pay-per-use license policy. Most of the organizations use this model which helps in managing load spikes, host SaaS applications, utilize interim infrastructure for developing and testing applications. Examples: - Google is an example of Public Cloud. Hybrid: - Hybrid Cloud is the combination of both Public and Private Clouds. This model is used for handling cloud bursting, where Private Cloud is not able to handle load spikes and require a fallback option as to handle the load. Example:- Microsoft Azure and Force.com are two examples of Hybrid Cloud. Community Cloud: - In this model, the Cloud infrastructure is shared by several organizations with the same policy. This model is cost efficient as compare to Private Cloud. As we know, State-level government departments where there is need to access the same data relating to the local population to infrastructural information such as road, hospitals etc., can use a Community Cloud to handle data and applications [1]. 1.3 Issues in Cloud Computing Cloud Computing facing a number of security issues. It is the responsibility of the service provider to satisfy the customer regarding data loss or data theft. Some of the problems [5] which are faced by the Cloud Computing are:

Page 197

International Journal of Emerging Trends & Technology in Computer Science (IJETTCS)


Web Site: www.ijettcs.org Email: editor@ijettcs.org, editorijettcs@gmail.com Volume 3, Issue 1, January February 2014 ISSN 2278-6856
Date Integrity: - Any client can access any data at any location. There is no differentiate between important data and common data, thus any one can access important data. So there is lack of data integrity in Cloud Computing. Data Theft: - As the data is stored on external servers because of cost effectiveness and flexibility reasons. So, there is higher possibility that the data can be stolen from the external server by malicious user. Privacy issues: - It is the responsibility of the vendor to make clients date secure from malicious users. As most of the servers are external, so vendor should be aware who is the owner of the data and who is maintaining the user, so that data can be secured. Data loss: - Data loss is also a big issue for customer. Due to financial or other problems, if vendor closes, there will be loss of data for the customer. Data Location: - When data is stored on the server, customer does not know where the data is stored. The Vendor does not reveal the location of data stored. The data might be stored in anywhere in the world. Security on vendor level: - Data is stored on external servers, so vendor should make sure that the server is secured where the data is stored and safe from all malicious users. Security on user level: - Even though the vendor satisfies the customer by providing good security, so now its the responsibility of the customer to safe their data from other users. to smooth distribution when the load of requests vary a lot [2]. D. Proposed Algorithm Proposed algorithm works on dynamic scheduling algorithms module. In cloud computing, machines are with different configuration as per the requirement of the user. In proposed algorithm, machines are arranged in descending order as per the RAM. Whenever request arrives for execution, VM with higher RAM is given the task with the condition that it is currently not busy. Otherwise VM next in order satisfying the same condition is returned. It also maintains a data structure that contains the associated weight count and the number of requests currently allocated to the VM. In this proposed algorithm works.

3. SIMULATION SETUP PERFORMANCE ANALYSIS

AND

Simulation setup and the performance of the different load balancing algorithms are done by using cloud analyst simulator. Users are divided on the world map into 6 regions shown in the table as well as the parameters of the Data Centers are shown in the tables [6]. Table 1:- User Based configuration User Base Region UB1 UB2 UB3 UB4 UB5 UB6 0 1 2 3 4 5

2. LOAD BALANCING ALGORITHMS


A. Round Robin Load Balancer (RRLB) In this, a number of requests are assigned by datacenter to a list of VMs on a rotating basis [2]. The first request is assigned to a VM- selected randomly from the list of VMs and then the DataCenter controller assigns the particular requests in a circular order [4]. Once the VM is assigned the request, the VM id is moved to the end of the list. In this manner, Round Robin Load Balancer works [3]. B. Throttled Load Balancer (TLB) In TLB, a record of the state of each virtual machine (busy/ideal) is maintained. If a request arrived regarding the allocation of virtual machine, ID of ideal virtual machine is send by TLB to the data center controller and data center controller allocates the ideal virtual machine [2]. C. Least Connection Scheduling Algorithm The least-connection scheduling algorithm [2] directs network connections to the server with the least number of established connections. This is one of the dynamic scheduling algorithms; because it needs to count live connections for each server dynamically. For a virtual server that is managing a collection of servers with similar performance, least-connection scheduling is good Volume 3, Issue 1 January February 2014

Data Center DC1 DC2 DC3

Table 2:- Data Center Configuration VMs Memory Bandwidth

10 10 10

1024 2048 4096

1000 1000 1000

4. PERFORMANCE ANALYSIS RESULTS


Algorithms Average Time (ms) Minimum Time (Ms) Maximum Time (ms) 369.14 369.14 369.27 369.14

Round Robin 300.22 237.12 ESCE 300.19 237.07 LCSA 300.16 237.07 Proposed 300.16 237.07 Alg. Overall Response Time (ms)

Page 198

International Journal of Emerging Trends & Technology in Computer Science (IJETTCS)


Web Site: www.ijettcs.org Email: editor@ijettcs.org, editorijettcs@gmail.com Volume 3, Issue 1, January February 2014 ISSN 2278-6856

300.22 300.21 300.2 300.19 300.18 300.17 300.16 300.15 300.14 300.13 Average Time RR ESCE LCSA TA

0.51 0.5 0.49 0.48 0.47 0.46 0.45 0.44 0.43 0.42 Average Time
ROUND ROBIN DC VM TIME ID (ns) DC1 0 12831 DC1 1 4491 DC1 2 3207 DC1 3 3208 DC2 0 3208 DC2 1 3850 DC2 2 3208 DC2 3 3208 DC3 0 2566 DC3 1 2566 DC3 2 5133 DC3 3 4492 ESCE DC DC1 DC1 DC1 DC1 DC2 DC2 DC2 DC2 DC3 DC3 DC3 DC3 LCSA DC DC1 DC1 DC1 DC1 DC2 DC2 DC2 DC2 DC3 DC3 DC3 DC3 PA DC DC1 DC1 DC1 DC1 DC2 DC2 DC2 DC2 DC3 DC3 DC3 DC3

RR ESCE LCSA TA

Data Center
Algorit hms

DC1

DC2

DC3

Av g

Min

Ma x

Av g

Mi n

Max

Avg

Min

Max

RB ESCE

0.3 71 0.5 03 0.3 69 0.3 85

0.03 4 0.06

0.7 58 27. 00 4 0.7 53 0.6 37

0.5 1 0.4 85 0.6 09 0.4 76

0.0 34 0.0 59 0.0 6 0.0 59

1.25 8 0.99 9 18.0 06 0.76

0.677 0.436

0.20 9 0.02 2 0.03 4 0.06

23.00 7 7.501

VM ID 0 0 0 0 0 0 0 0 0 0 0 0

TIME (ns) 68651 8341 6416 6416 10907 7058 7057 7058 10266 7058 7058 7057

VM ID 0 0 0 0 0 0 0 0 0 0 0 0

TIME (ns) 71217 7699 5775 5774 9624 5775 6416 5775 8983 5774 5774 5774

VM ID 0 0 1 1 0 0 0 0 0 1 2 3

TIME (ns) 87898 8982 8983 8340 8983 7057 6416 6416 8340 8341 8982 10907

Resource Allocation time(ns)

LCSA PA

0.02 2 0.02 2

0.379 0.478

0.883 0.761

5. CONCLUSION
The efficient load balancing techniques provide performance benefits to the user. By introducing a new load balancing algorithm, which can dynamically adapt to situation on servers to which they actually forward traffic, most of the traffic related problems can be avoided. Proposed algorithm tested with the help of three areas: allocation time, responsiveness, and efficiency. Proposed Load Balancing algorithm works efficiently on these three parameters. In future, work can be done by considering others parameters to make load balancing more efficient and effective.

Data Center Request Servicing Time(ms)

Algorithms

Userbas e

Average time(ms)

Minimum time(ms)

Maximum time(ms)

RB ESCE LCSA PA

UB1 UB1 UB1 UB1

300.221 300.189 300.161 300.161

237.119 237.069 237.069 237.069

369.14 369.14 369.265 369.14

References /Bibliography
[1] Ajay Gulati1, Ranjeev.K.Chopra2 Dynamic Round Robin for Load Balancing in a Cloud Computing IJCSMC, Vol. 2, Issue. 6, June 2013, [2] Amjad Mahmood and Irfan Rashid, Comparison of Load Balancing Algorithms for Clustered Web Servers, 5th International Conference on IT & Multimedia at UNITEN(ICIMU 2011) Malaysia, 1416 November 2011 [3] Ankush P. Deshmukh, Prof. Kumarswamy Pamu, Applying Load Balancing: A Dynamic Approach, International Journal of Advanced Research in Computer Science and Software Engineering(IJARCSSE), Volume 2, Issue 6, June 2012,. [4] Bashir Alam, M.N. Doja, R. Biswas, Finding Time Quantum of Round Robin CPU Scheduling Page 199

Response Time by Regions(ms)

Algorithm

Average Time(ms) 0.51 0.47 0.45 0.45

Minimum Time(ms) 0.03 0.02 0.02 0.02

Maximum Tim(ms) 23.01 27.00 18.01 0.76

RB ESCE LCSA PA

Data Center Processing Time(ms) Volume 3, Issue 1 January February 2014

International Journal of Emerging Trends & Technology in Computer Science (IJETTCS)


Web Site: www.ijettcs.org Email: editor@ijettcs.org, editorijettcs@gmail.com Volume 3, Issue 1, January February 2014 ISSN 2278-6856
Algorithm Using Fuzzy Logic, 2008 International Conference on Computer and Electrical Engineering, pp. 795-798, November 2009. [5] Branko Radojevi, Mario agar, Analysis of Issues with Load Balancing Algorithms in Hosted (Cloud) Environments, MIPRO 2011, May 23-27, 2011, Opatija, Croatia. [6] Jaspreet kaur, Comparison of load balancing algorithms in a Cloud, International Journal of Engineering Research and Applications(IJERA), Vol. 2, Issue 3, pp. 1169-1173, May-June 2012. [7] Hendra Rahmawan, Yudi Satria Gondokaryono, The Simulation of Static Load Balancing Algorithms, International Conference on Electrical Engineering and Informatics, pp. 640-645, 5-7 August 2009. [8] Rudra Koteswaramma, Client-Side Load Balancing and Resource Monitoring in Cloud, International Journal of Engineering Research and Applications (IJERA), Vol 2, Issue 6, pp. 167-171, Nov-Dec 2012. [9] Sharada Patil, 2Prof. Dr. Arpita Gopal, Cluster performance evaluation using load balancing algorithm, Information Communication and Embedded Systems (ICICES), pp. 104-108, 21-22 Feb. 2013. [10] Soumya Ray and Ajanta De Sarkar, Execution analysis of load balancing algorithms in cloud computing environment, International Journal on Cloud Computing Services and Architecture (IJCCSA),Vol.2, No.5, pp. 1-13, October 2012. [11] Venubabu Kunamneni, Dynamic Load Balancing for the Cloud, Internationl Journal of Computer Science and Electrical Engineering (IJCSEE), Vol. 1, Issue 1,2012.

AMRITPAL SINGH is a student of M.Tech (CSE) at BCET, Gurdaspur, did his graduation in Information technology from BCET, Gurdaspur. His current research area of interest is Cloud Computing. SUNIL KUMAR GUPTA is an Associate Professor in the Department of Computer Science & Engineering at BCET, Gurdaspur. His current research area of interest is distributed computing.

RAJEEV KUMAR BEDI is an Assistant Professor in the Department of Computer Science & Engineering at BCET, Gurdaspur. His current research area of interest is cloud computing.

Volume 3, Issue 1 January February 2014

Page 200

Das könnte Ihnen auch gefallen