So I am using Paralution to quickly check the performance of different solvers on some test problems of increasing size. I also decided to test every solver in CSR and DIA format to check whether DIA indeed does better (I work with diagonally structured matrices). Now it happens that for my larger test problems, BiCGStab does perfectly fine when doing in CSR format, but diverges every single time when using DIA. For the smaller problems it works fine with both CSR or DIA. Any idea what could be the reason for this? The storage format of the matrix shouldn’t affect the behavior of the solver, except for the computing time, or did I understand this wrongly?
sorry for the delay. Due to different rounding, small variations could happen between CSR and DIA format. However, this should not lead to divergence. Is it possible to get access to your linear system where this is happening?