Increasing SDN Network Performance Using Load Balancing Scheme on Web Server

In this paper, we propose to test Least Connection and IP Hash load balancing algorithm to be applied to web servers over SDN networks. The use of load balancing on web servers over SDN networks began to attract many researchers. The problem of using appropriate load balancing algorithms for this case is still debatable. One of the problems is determining the correct algorithm for web server which has bound function (session). Load balancing algorithm that can provide stable performance is a static algorithm. However, this algorithm does not have a function bound to a server, so it is not possible to apply to a web server with bound functions, such as login session function. Therefore, a possible algorithm was developed for the function Two examples of algorithms that have the potential to be an effective algorithm for the case of this web server is the Hash and Least Connection IP algorithm. We have implemented and tested the load balancing performance using Least Connection and IP Hash algorithm. In this test, both algorithms will be tested with some crucial parameters for a web server. These parameters are Response Time, Throughput and Resource Utilization. The result of the test with the parameters is IP Hash algorithm gives performance response time 17% more optimal, 10% more optimal throughput and 8% usage memory more efficient than Least Connection algorithm.

[1]  Timotius Witono,et al.  ANALISIS ALGORITMA ROUND ROBIN, LEAST CONNECTION, DAN RATIO PADA LOAD BALANCING MENGGUNAKAN OPNET MODELER , 2016 .

[2]  Yoohwan Kim,et al.  Hash-based Internet traffic load balancing , 2004, Proceedings of the 2004 IEEE International Conference on Information Reuse and Integration, 2004. IRI 2004..

[3]  Japinder Singh,et al.  Implementation of Server Load Balancing in Software Defined Networking , 2016 .

[4]  Xiao Guo,et al.  SDN-based load balancing strategy for server cluster , 2014, 2014 IEEE 3rd International Conference on Cloud Computing and Intelligence Systems.

[5]  Jim Esch,et al.  Software-Defined Networking: A Comprehensive Survey , 2015, Proc. IEEE.