Conformance verification for the SEDRIS implementation
A fundamental discriminator for level of conformance verification is the numerical difference between a result obtained from the mathematical specifications in the SRM and the corresponding result obtained by an implementation.
- This requires that there be a Reference Implementation.
- A data set produced by the Reference Implementation to support conformance testing is called a Reference Data Test Set.
- For this purpose, the data points obtained from the RDS are taken to be the exact or true data points.
Quantities in the Reference Data Test Set are called test points.
- Each implementation to be tested shall have provisions for recording the information needed to compare to the Reference Data Test Set to support of conformance testing
- This set of output data is called a Test Data Set.
- The difference between the data points of an Reference Data Test Set and the Test Data Set of a particular implementation is referred to as an error.