A Framework for Resilient Computer Networks Dependability in Universities

Thesis Title: A Framework for Resilient Computer Networks   Dependability in Universities.

Student‘s Name: Omasaja Ekadakada James

Supervisors‘ Names:

1. Betty Mayeku.
2. Daniel Otanga.

Abstract:

Computer networks bring along many benefits in the present day society. However, full benefits of computer networking are yet to be realized because of non-resilience issues. One of the sectors that has been a forefront user of computer networks and is yet to tap into its full benefits is universities. Although there are some techniques that have been explored to address computer network resilience issues universities pose a unique challenge due to the diverse conditions they present. The purpose of this study was to examine computer networks and develop a framework for computer network resilience in achieving computer network dependability in the context of universities. The study was guided by the following objectives; establishment of technologies currently in use, determination of techniques currently in use to achieve computer network resilience and development of a framework for computer network dependability in universities. The study adopted a survey research design targeting universities where Kibabii University was purposively sampled.  The 6246 undergraduate, 248 masters and 42 PhD students and 430 computer network using staff were target population. Random and purposive sampling were used where appropriate. The study used content analysis, questionnaires, interviews for data collection and a focus group for validation process. Data collection tools were given to 3 experts to validate where they scored a validity value of 78.3%. Inferential and descriptive statistics were used to analyse the data. It was established that 75.2% of respondents had access to the studied network. 66.7% were positive on hotspots, 72.1 % appreciated the distribution of hotspots in campus, 52.2% acknowledged reliability of the network, 50.1% used cloud services and 50.2% reported network outages were on the minimum. Ethernet, switches and routers, Wi-Fi and network security hardware and software were found as current computer network building technologies. Redundancy, load balancing, network segmentation, disaster recovery planning, regular backups and network monitoring were determined as computer network resilience techniques currently in use.  A KMO value of 80.3% (0.803) justified use of factor analysis on the collected data. Network, connectivity and human characteristics with factor loadings of (0.1898), (0.4359) and (0.3243) were the main constructs of the framework. The researcher hopes that focusing on connectivity, human behavior and network characteristics in that order will guarantee network end users a worthwhile experience, help computer hardware and software vendors fine tune their solutions and guide governments and universities plan and invest in enlightened environments when investing in computer networks especially in universities.The researcher recommends that further studies be carried out to establish the efficay and use of the network based on the core network attributes in non-education based institutions and that more studies be carried out on the nature of network connection and the user attitudes in use and adoption.