=================================== = sparse eigenanalysis statistics = =================================== number of equations = 17262 no. of nonzeroes in lower triangle of a = 451507 number of compressed nodes = 2913 no. of compressed nonzeroes in l. tri. = 11540 amount of workspace currently in use = 2921243 b has general form. no. of nonzeroes in lower triangle of b = 452232 max. amt. of workspace used = 9543677 no. of nonzeroes in the factor l = 3048624. number of super nodes = 1116 number of compressed subscripts = 102840 size of stack storage = 422910 maximum order of a front matrix = 666 maximum size of a front matrix = 222111 maximum size of a front trapezoid = 40544 no. of flt. pt. ops for a single factor = 8.849281D+08 no. of flt. pt. ops for a block solve = 1.228081D+07 no. of flt. pt. ops for block mtx mult. = 3.646580D+06 total no. of flt. pt. ops for factor = 1.769856D+09 total no. of flt. pt. ops for solve = 6.631635D+08 total no. of flt. pt. ops for mtx mult. = 2.297345D+08 actual no. of nonzeroes in the factor l = 3048624. actual number of compressed subscripts = 102840 actual size of stack storage used = 484437 pivot tolerance used for factorization = 0.000000 factorization panel size = 64 factorization update panel size = 32 solution block size = 8 Lanczos block size = 2 total number of factorizations = 2 total number of lanczos runs = 2 total number of lanczos steps = 25 total number of block solves = 27 time (cpu & wall) for structure input = 0.062401 0.711744 time (cpu & wall) for ordering = 0.031200 0.031892 time (cpu & wall) for symbolic factor = 0.000000 0.011726 time (cpu & wall) for value input = 0.093601 0.775755 time (cpu & wall) for numeric factor = 0.561604 0.292395 computational rate (mflops) for factor = 3151.433295 6052.972179 time (cpu & wall) for numeric solve = 0.499203 0.233415 computational rate (mflops) for solve = 1328.445312 2841.133319 time (cpu & wall) for matrix multiply = 0.124794 0.108695 computational rate (mflops) for mult. = 1840.910046 2113.578092 cost (elapsed time) for sparse eigenanalysis ----------------------------- lanczos run start up cost = 0.025078 lanczos run recurrence cost = 0.221852 lanczos run reorthogonalization cost = 0.200502 lanczos run internal eigenanalysis cost = 0.000502 lanczos eigenvector computation cost = 0.219292 lanczos run overhead cost = 0.003634 total lanczos run cost = 0.670860 total factorization cost = 0.292404 shift strategy and overhead cost = 0.732679 total sparse eigenanalysis cost = 1.695944 i/o stats: unit file length amount transferred words mbytes words mbytes ---- ----- ------ ----- ------ 21 205680. 2. Mb 617040. 5. Mb 25 552384. 4. Mb 4211928. 32. Mb 28 517860. 4. Mb 3245256. 25. Mb 11 69048. 1. Mb 69048. 1. Mb Total: 1344972. 10. Mb 8143272. 62. Mb