How cloud intelligence dashboard analyse forecast values?

Cloud Intelligence Dashboards use Quicksight's ML powered forecasting. By clicking the menu button on a visualization and selecting "Edit Forecast" you can view and adjust the properties. Here are some resources: https://docs.aws.amazon.com/quicksight/latest/user/forecasts-and-whatifs.html https://docs.aws.amazon.com/quicksight/latest/user/forecast-function.html

The Cloud Intelligence Dashboards, specifically the Cost and Usage Dashboards Operations Solution (CUDOS), provide forecasted values for AWS costs and usage. While the exact forecasting methodology is not explicitly detailed in the provided information, we can infer some aspects of how these forecasts are generated based on AWS's general approach to cost forecasting.

AWS Cost Explorer, which likely feeds data into the Cloud Intelligence Dashboards, uses historical cost and usage data to generate forecasts. When you first sign up for Cost Explorer, AWS prepares data about your costs for the current month and the last 13 months, and then calculates a forecast for the next 12 months. This historical data is used as a basis for projecting future costs.

The forecasts typically include both a point estimate and a range (lower and upper bounds). These bounds likely represent a confidence interval around the point estimate, taking into account factors such as historical variability in costs, seasonality, and trends in your usage patterns.

While a specific forecasting service isn't mentioned, AWS likely uses internal algorithms and statistical methods to generate these forecasts. These could include time series analysis techniques, regression models, or more advanced machine learning algorithms that can account for various patterns and trends in your historical data.

It's important to note that the forecasts are updated regularly, with Cost Explorer refreshing cost data at least once every 24 hours. This means the forecasts you see in your Cloud Intelligence Dashboard are continuously adjusted based on your most recent cost and usage data.

The lower and upper bounds you observe are likely calculated to provide a range of potential outcomes, considering the inherent uncertainty in forecasting. These bounds help in understanding the potential variability in future costs, allowing for more informed budgeting and planning decisions.
Sources
Use Landing Zone Accelerator on AWS customizations to deploy Cloud Intelligence Dashboards | AWS Public Sector Blog
Analyzing your costs and usage with AWS Cost Explorer - AWS Cost Management