CoreLB: towards providing multi-metric score-based load balancing using OpenFlow and SNMP
by Rodrigo Da Rosa Righi; Cristiano André Da Costa; Clebio Dossa; Felipe Rabuske; Diego Luis Kreutz
International Journal of Information Technology, Communications and Convergence (IJITCC), Vol. 3, No. 3, 2019

Abstract: Load balancing is still one of the most crucial services offered by service providers. In spite of having several load balancing solutions in the market, this area has gained attraction with SDN and OpenFlow. In this article, we present CoreLB, a new load balancing model that takes into account a broader range of metrics from both networking devices and servers to achieve a better balancing of the servers' workload. CoreLB combines statistics of servers, using the SNMP protocol, and forwarding devices, through the OpenFlow protocol. Its scientific contribution is twofold: 1) a novel score-based load balancing metric; 2) a service agnostic metric, i.e., a metric capable of handling the requests of CPU-bound, network-bound and IO-bound services. We evaluated CoreLB through a prototype implementation, which uses web services and handles different types of requests and load balancing scenarios. Our results show that CoreLB provides a better response time from the user's viewpoint. In average, the request processing time has been reduced by 19%, when compared to non-balancing, and 9% when compared other load balancing strategies. In addition, we also achieve an evener resource utilisation when using CoreLB.

Online publication date: Tue, 14-Apr-2020

The full text of this article is only available to individual subscribers or to users at subscribing institutions.

 
Existing subscribers:
Go to Inderscience Online Journals to access the Full Text of this article.

Pay per view:
If you are not a subscriber and you just want to read the full contents of this article, buy online access here.

Complimentary Subscribers, Editors or Members of the Editorial Board of the International Journal of Information Technology, Communications and Convergence (IJITCC):
Login with your Inderscience username and password:

Username:        Password:          Forgotten your password?

Want to subscribe?
A subscription gives you complete access to all articles in the current issue, as well as to all articles in the previous three years (where applicable). See our Orders page to subscribe.

If you still need assistance, please email subs@inderscience.com