A study of efficient concurrent integration methods of B-Spline basis functions in IGA-FEM

Based on trace theory, we study efficient methods for concurrent integration of B-spline basis functions in IGA-FEM. We motivate the discussion with heat-equation problem discretized forward Euler time-marching scheme time and IGA-FEM space, leading to cost associated assembling Mass matrix. consider several scenarios parallelization two standard methods; classical one sum factorization. aim efficiently utilize hybrid memory machines, such as modern clusters, by focusing non-obvious layer shared part concurrency. estimate performance computations a GPU provide strategy performing practical implementations.