Network Automation in a Multi-vendor Environment

• Main Author: Larsson, Jonas
• Format: Others
• Language: Swedish
• Published: Luleå tekniska universitet, Institutionen för system- och rymdteknik 2020
• Subjects: Computer Sciences; Datavetenskap (datalogi)
• Online Access: http://urn.kb.se/resolve?urn=urn:nbn:se:ltu:diva-80810

Automation is something that effects every occupation, and everyone wants more of. Networking is not any different were the systematic tasks that can be prone to human errors and the long time-wasting configuration surely could be automated. But a problem in networking is that when there aren’t any standards to follow vendors do their own solutions that only works on their own devices. This is not a problem if you're only using one vendor in your network, but what if you don't then how do you automate your network? This is the question this report is trying to answer. First by trying to develop a new solution in Python to automate networking tasks and then see if there is any existing product that works in a multi-vendor network that can be tested. Then test the programs, see what functions exist and demonstrate how they work and see advantages and disadvantages with the solutions. The new program in Python is planned, the different functions that the program is going to have is explained, the development is done and then tested. Then a program called UNIMUS is installed and tested. The development resulted in a program that’s called Kraft Prime. The testing on Kraft Prime shows a program that offers automation on different tasks like configuration and upgrading. The results from testing UNMIUS also shows a program that offer automation functions like mass configuration and configuration backups. The results from the development and testing shows that it is possible to develop your own network automation program in Python that offers different kinds of automation in a multi-vendor environment. But also, that there are existing programs like UNIMUS that also offers automation. But both solutions come with their ups and downs that must be considered before implementing any of the solutions.