dc.description.abstract |
Uganda as a developing country has been having some of the organisations and companies getting their services accessible online. These organisations include UMEME, URA, National Water. KCCA and PayWay with services used on a daily basis by Ugandans. The services include electricity, tax, water and financial payment.
At peak hours when many users are making payment or accessing the services, their network goes off at times. The causes network outage range from a number reasons including link failure, server failure, network device failure or wrong configuration on a device. Some of the causes are natural and unavoidable like broken cables, wrong configurations and others can. be mitigated.
Among the causes, is network device failure and yet all the devices provide different functionality for the services. The devices include servers for storage and handling requests from user: routers for routing packets within a network, switches for layer 2 traffic flow and gateways for providing a boundary between an enterprise network and the Internet Service- Provider (ESP).
This project focuses on the gateway among devices mentioned above because it is the highway for all traffic flowing in and out of the network and so being more susceptible to bottlenecking compared to any other device. When many users ate using the service, interfaces on 'the gateway become over utilised resulting into a delay a response hence poor delivery of a service. Hence the need for optimal selection of an interface depending on the type traffic flowing through network.
In incidents of high interface utilisation, protocols like Link Aggregation Protocol (LAgP) and Port Aggregation Protocol (PAgP) are used to create port channels with higher throughput. Also optical fibre which has a high bandwidth can be used to accommodate more traffic. However, both need extra hardware to be implemented in a network which implies more cost to implement.
The objective of this project is to develop a python application that monitors interfaces on gateways running Host Standby Redundancy Protocol between them and automatically send traffic to an underutilised interface on the gateway. The application implements a policy to send specific traffic through a less utilised Vlan connected to one of the gateways. This improves packets per second flowing to and the host, hence an improvement in the reliability of the network and the service. |
en_US |