Internal Data Processing

CAVE processes multiple results in one analysis in two dimensions:

• parallel runs denotes multiple runs with the same preliminaries, scenarios and only different results

• budgets (aka fidelities) denotes runs with different approximations of the target problem.

Internally, CAVE stores a ConfiguratorRun for each parallel run. ConfiguratorRuns are stored in a RunsContainer, which provides easy methods for aggregation or reduction of ConfiguratorRuns.