The following figure shows a high-level workflow of everyday user tasks and how data flows through Kyvos.
Kyvos user roles and workflows
A person may do more than one of these roles.
- Creates connections to data and Query Engines
- Registers sources such as Hadoop, SQL, other files or tables for use with Kyvos
- Manages users and security
- Monitors Kyvos performance
- Imports and exports Kyvos data
- May apply some transformation logic to the data
Creates datasets which are collections of data used for analysis
- Uses formulas, filters, and other transformation tools to do precalculations on that data–to speed up subsequent analysis and highlight the important data
- Optionally, creates relationships within the data and between different data sources. For example, defining a relationship between a field in one file with an associated field in another file
- Designs and builds cubes for multidimensional analysis
- Creates workbooks and worksheets
- Creates interactive dashboards, charts, filters data
- Slices and dices into the data to develop insights
- Exports chart data to share with others
- Uses Kyvos data with external analysis tools
Why create datasets and cubes
Why do I need to create datasets and cubes?
Big data often comes from a variety of sources, and you need to organize it in some way so you can work with it effectively.
Datasets and cubes allow you to quickly and efficiently work with big data. Datasets enable you to transform the data before you use it. For example, you can group and filter data, pre-calculate aggregations of data, and hide data that is part of the source file but that you are not going to use during data analysis. Create relationships between data from different sources so that you can use them together for analysis. You can even specify how data is formatted, such as dates and currency.
Note that you can also create queries using raw data that is not pre-aggregated into a dataset or cube.