Three major cloud providers and their applicability for hosting a chatbot
Hakkarainen, Jarmo (2017)
Hakkarainen, Jarmo
Haaga-Helia ammattikorkeakoulu
2017
All rights reserved
Julkaisun pysyvä osoite on
https://urn.fi/URN:NBN:fi:amk-2017120720181
https://urn.fi/URN:NBN:fi:amk-2017120720181
Tiivistelmä
Up until this project, Liquid Studio has mainly used AWS. To avoid tying everything to one platform branching out was required. To test out the capability and viability of both Microsoft Azure Cloud and Google Cloud Platform a chatbot would be deployed on each platform to map out the process and requirements. The chatbot was chosen due to its dependency on external APIs, thus demonstrating a multi-cloud solution. A chatbot is in essence a software capable of having an automated conversation with the user.
The key metric used was the need for refactoring, the less the better with an added consideration to time spent on the deployment and configuration. The end goal was to have the chatbot running on all platforms.
The plan was to keep the code base the same, and only modify the scripts required for deployment. This was achieved sufficiently since only a few scripts were modified and a few files added. Resources were chosen according to documentation available on each platform and best practices in the field.
The project was timed for three months. Tooling used in the thesis was the same development setup that is generally used within Liquid Studio. This setup includes Visual Studio Code as the code editor and a MacBook Pro as a workstation. The default terminal was used for console access and related work.
The key metric used was the need for refactoring, the less the better with an added consideration to time spent on the deployment and configuration. The end goal was to have the chatbot running on all platforms.
The plan was to keep the code base the same, and only modify the scripts required for deployment. This was achieved sufficiently since only a few scripts were modified and a few files added. Resources were chosen according to documentation available on each platform and best practices in the field.
The project was timed for three months. Tooling used in the thesis was the same development setup that is generally used within Liquid Studio. This setup includes Visual Studio Code as the code editor and a MacBook Pro as a workstation. The default terminal was used for console access and related work.