Developing AI solutions for Cloud Cost Management by using open data made available by AWS.
Most companies nowadays are paying customers of one of the many cloud vendors in the industry, or are planning to be. These cloud providers keep changing their prices from time to time. However, a lack of information about how and when these prices change results in a lot of uncertainty for customers. Being able to understand price changes would help customers take appropriate measures to best manage their costs. Hence, given a dataset of cloud price lists, we aim to build a Cost-Optimization model that allows the user to make the best decision on how cloud services should be managed over time.
In order to be able to make intelligent decisions about costs and resources, it is necessary to understand not only the current state of the resources, but also what will be the possible evolution of them. The customer should be capable of understanding the future evolution of the resources and costs if no action is taken, or if some action is taken on the source.
In this project, we aim to manage Cloud Costs by using Machine Learning algorithms such as Classification and Linear Regression. Besides choosing the right cloud service and service provider, we aim to forecast the upcoming costs and hence be prepared for any overhead and cut on extra costs if things go out of hand. Finally, we aim to optimize cloud usage by finding adjacencies in cloud services to choose the best provider.
Below is a video overview of the Project Statement and Exploratory Data Analysis in the JupyterHub environment.
video: https://www.youtube.com/watch?v=jne4auRnxJ4&feature=youtu.be