Dear Developers,<br><br>I tried to following optimized matrix data setup for a block-sparse mat solver but when I run the code, too much time is taken "just" for the first step of SNES procedure (about 20 to 30 times than the next steps ?! ).<br>
<br>I did above as follows:<br><br>1- Allocate matrix( matrices ):<br><br>MatSetType ( JAC, MATSEQBAIJ );<br>MatSetSizes ( JAC, PETSC_DECIDE, PETSC_DECIDE, nt, nt ); <br>MatSeqBAIJSetPreallocation ( JAC, tot, 0, nnz );<br>
MatSetFromOptions ( JAC );<br><br>where nnz[] is account regarded to connectivity<br><br>2- Matrix Data calculation:<br><br>MatSetValues( JAC, tot, IDX, tot, JDX, VAL, ADD_VALUES );<br><br>where IDX, JDX, VAL are blocks of data with size tot.<br>
<br>3- Did the same for preconditioning matrix.<br><br>4- used PCILU for preconditioning.<br><br>I also print out log_summary but I actually can't find out if I set the memory optimized ( I don't no how to follow the logfile and check what parameters ).<br>
<br>Please let me know how would be the failure in my procedure...<br><br>Thanks a lot,<br>BehZad<br><br>