Data granularity is a measure of the level of detail in a data structure. In time-series data, for example, the granularity of measurement might be based on intervals of years, months, weeks, days, or hours. For ordering transactions, granularity might be at the purchase order level, or line item level, or detailed configuration level for customized parts. The name field could represent the full name or have separate entries for first name, middle name, and last name.
The level of data granularity determines what analysis can be performed on the data, and whether results from that analysis lead to appropriate conclusions. In the last example, name matches may not be the same person unless (and even if) suffixes are considered. The more granularity, the more information is available for analysis, but at a cost of increased storage, memory, and required computing resources. Some analysis may require information to be analyzed at a higher level, which would require aggregating the underlying detail into the higher level of granularity.
The C3 AI® Suite handles big data with aplomb across its scalable, flexible, and elastic cloud computing infrastructure. Using a powerful expression capability, it is easy to aggregate detailed information at the right level for analysis, and to drill down to greater levels of granularity as needed. Data engineers can use the C3 AI Data Integrator to view the entire data model, drill down to individual fields and data values, and use visualizations to explore the history of values over time.