Benchmarking to close the credibility gap: A computational BioEM benchmark suite

The dearth of verification, validation, and performance benchmarks is identified as a roadblock to further progress in computational electromagnetics. The necessary ingredients for a useful benchmark suite are an application-specific list of problems, reference solutions, performance (error and computational cost) measures, and online databases publicizing comparisons. Computational cost comparisons are particularly difficult, rare, and important. As a case study, a benchmark suite for comparing existing and future computational bioelectromagnetics methods is developed.