Forecasting cost and usage with Amazon Forecast
Amazon Forecast is best suited for time-series data. Time-series data is data collected over a period that shows changes over time. This is well suited for CUR data since AWS collects your cost and usage data with a daily time interval. Amazon Forecast expects your data to have a timestamp field that aligns with the CUR schema since the cost and usage line-item column has this format already.
Important note
Amazon Forecast has costs associated with its usage. Strongly consider whether using an artificial intelligence (AI)-based service is worth the investment since ML-related costs tend to be more expensive than storage, compute, and networking hourly costs. I present this solution as an option but it may not be suitable for everyone.
To get started, access the Amazon Forecast console page and create a dataset group. This is the first step since you need to define what your time-series data will be for forecasting. Provide a...