Soeren,I am forwarding the tests to the dev list so that it gets into the archive and everybody can check it out.
For everybody - this relates to the discussion about the NR lineqs solver replacement,
Helena Helena Mitasova Dept. of Marine, Earth and Atm. Sciences 1125 Jordan Hall, NCSU Box 8208, Raleigh NC 27695 http://skagit.meas.ncsu.edu/~helena/ Begin forwarded message:
From: Sören Gebbert <[EMAIL PROTECTED]> Date: May 6, 2007 6:19:03 AM EDT To: Helena Mitasova <[EMAIL PROTECTED]> Subject: Re: LINEQS Helena, Thanks for the chapter. I have tested the solvers with several vector maps from the new demo location. The script which performs the tests is attached. So you are able to reproduce the tests. The log file, including time measurement and error log, is attached as well.I have created two simple bar charts. The first represents the processed vector maps and the cpu time consumption of each solver, measured by the time command. The second one represents the elev result error between the iterative solvers and the nrlu solverfor each processed vector map. All the data presented in the charts is available in the log file. The cg solver is the fastest solver, the new direct lu and gauss solver are 20% slower than the nrlu algorithm. The computation error of the iterative solvers is small enough to beaccepted. The gpde lu and gauss solver producing exactly the same result as the nrlualgorithm (see log file). So the new lu and the cg solver are the best replacement for nrlu. I did not tested parallel computation, because i have no multi core computer at home.But, the iterative solvers partially failed on vector maps with low data density (contour linesin maps elev_contour_3m and elev_ned10m_cont10m),in case the npmin was set to 50 or lower. A value of 50 is not meaningful for the maps,but its the users choice ... . So much more testing is needed. Anyway, the gpde lu solver can handle this. Important region and v.surf.rst settings: g.region rural_1m res=2 npmin=500 Best regards Soeren Helena Mitasova schrieb:On May 5, 2007, at 4:00 PM, Sören Gebbert wrote:yes - actually v.surf.rst (s.surf.tps) originally looked like v.vol.rst which is a directHi, now i know the source of the linear equation solver which you had send me. Its is the solver implemented in v.vol.rst in user4.c. :Drewrite of our original fortran code. There should also be v.volt.rst that interpolates in 3D+time space.I really wanted all 3 programs and r.resamp.rst to be based on a single library asmost of the code is almost the same, only the dimensions change, but Irina (one of the russian twins that worked with me at CERL and did a lot of coding for GRASS, you can see Olga and Irina mentioned throughout various GRASS modules)managed to get it to where it is now by the time she finished her Masters and CERL dropped the GRASS project. So it is a quite ugly code right nowbut it works . I doubt that anybody would have time to rewrite it.Looks like good old fortran code converted to C. I think i will patch v.vol.rst too.yes that would be good. I am thinking that it may be useful to submitthe options for solvers that you have added into CVS, I would like to let people from GEON project try it out too (http://lidar.asu.edu/), especiallythe parallel processing.It would be great if you could try out the performance of the solvers with the new data set - both in terms of how fast they are and whether there are any differences in the results. The data are here, (set the region to rural_1m)http://mpa.itc.it/grasstutor/data_menu3rd.phtmlTry it with elev_lid792_bepts, elev_lid792_mrpts, elev_lid792_randpts and elev_contour_3m. I have attached chapter6.tex file that has all the commands so you can just cut and paste them into a script and run it all at once(ignore the figures in the pdf file - they are still for sperafish).These are all small files so a few tests with millions of points would be useful too (e.g. the Jockey's ridge data from the second edition available at the same web site). That should help to decide which method put as default and which are not needed (to make the life easier for the user). I will then get Jaro and some other people test it more to make sure that the new solvers work for everybody's data.HelenaBest regards Soeren
RST_solver_test.sh
Description: application/shellscript
########################################################## ++ Testing solver nrlu with map elev_lid792_bepts647.86user 4.10system 10:52.79elapsed 99%CPU (0avgtext+0avgdata 0maxresident)k0inputs+0outputs (0major+89167minor)pagefaults 0swaps ########################################################## ++ Testing solver lu with map elev_lid792_bepts853.22user 5.02system 14:19.62elapsed 99%CPU (0avgtext+0avgdata 0maxresident)k0inputs+0outputs (0major+89451minor)pagefaults 0swaps-- statistics of differences bewteen nrlu and lu solver based on elevationtotal null and non-null cells: 131250 total null cells: 0 Of the non-null cells: ---------------------- n: 131250 minimum: 0 maximum: 0 range: 0 mean: 0 mean of absolute values: 0 standard deviation: 0 variance: 0 variation coefficient: nan % sum: 0 -- statistics of differences bewteen nrlu and lu solver based on slope total null and non-null cells: 131250 total null cells: 0 Of the non-null cells: ---------------------- n: 131250 minimum: 0 maximum: 0 range: 0 mean: 0 mean of absolute values: 0 standard deviation: 0 variance: 0 variation coefficient: nan % sum: 0 ########################################################## ++ Testing solver gauss with map elev_lid792_bepts844.28user 5.10system 14:10.43elapsed 99%CPU (0avgtext+0avgdata 0maxresident)k0inputs+0outputs (0major+88890minor)pagefaults 0swaps-- statistics of differences bewteen nrlu and gauss solver based on elevationtotal null and non-null cells: 131250 total null cells: 0 Of the non-null cells: ---------------------- n: 131250 minimum: 0 maximum: 0 range: 0 mean: 0 mean of absolute values: 0 standard deviation: 0 variance: 0 variation coefficient: nan % sum: 0-- statistics of differences bewteen nrlu and gauss solver based on slopetotal null and non-null cells: 131250 total null cells: 0 Of the non-null cells: ---------------------- n: 131250 minimum: 0 maximum: 0 range: 0 mean: 0 mean of absolute values: 0 standard deviation: 0 variance: 0 variation coefficient: nan % sum: 0 ########################################################## ++ Testing solver cg with map elev_lid792_bepts298.31user 2.34system 5:01.01elapsed 99%CPU (0avgtext+0avgdata 0maxresident)k0inputs+0outputs (0major+89731minor)pagefaults 0swaps-- statistics of differences bewteen nrlu and cg solver based on elevationtotal null and non-null cells: 131250 total null cells: 0 Of the non-null cells: ---------------------- n: 131250 minimum: -1.52588e-05 maximum: 1.52588e-05 range: 3.05176e-05 mean: 9.53311e-09 mean of absolute values: 5.71057e-07 standard deviation: 2.25911e-06 variance: 5.10356e-12 variation coefficient: 23697.5 % sum: 0.0012512207 -- statistics of differences bewteen nrlu and cg solver based on slope total null and non-null cells: 131250 total null cells: 0 Of the non-null cells: ---------------------- n: 131250 minimum: -1.52588e-05 maximum: 1.52588e-05 range: 3.05176e-05 mean: 9.53311e-09 mean of absolute values: 5.71057e-07 standard deviation: 2.25911e-06 variance: 5.10356e-12 variation coefficient: 23697.5 % sum: 0.0012512207 ########################################################## ++ Testing solver bicgstab with map elev_lid792_bepts451.35user 3.36system 7:42.13elapsed 98%CPU (0avgtext+0avgdata 0maxresident)k0inputs+0outputs (0major+90562minor)pagefaults 0swaps-- statistics of differences bewteen nrlu and bicgstab solver based on elevationtotal null and non-null cells: 131250 total null cells: 0 Of the non-null cells: ---------------------- n: 131250 minimum: -3.8147e-05 maximum: 1.52588e-05 range: 5.34058e-05 mean: -6.01051e-08 mean of absolute values: 3.5133e-07 standard deviation: 1.90371e-06 variance: 3.62411e-12 variation coefficient: -3167.3 % sum: -0.0078887939-- statistics of differences bewteen nrlu and bicgstab solver based on slopetotal null and non-null cells: 131250 total null cells: 0 Of the non-null cells: ---------------------- n: 131250 minimum: -3.8147e-05 maximum: 1.52588e-05 range: 5.34058e-05 mean: -6.01051e-08 mean of absolute values: 3.5133e-07 standard deviation: 1.90371e-06 variance: 3.62411e-12 variation coefficient: -3167.3 % sum: -0.0078887939 ########################################################## ++ Testing solver nrlu with map elev_lidrural_mrpts427.92user 2.97system 7:11.51elapsed 99%CPU (0avgtext+0avgdata 0maxresident)k0inputs+0outputs (0major+87940minor)pagefaults 0swaps ########################################################## ++ Testing solver lu with map elev_lidrural_mrpts587.63user 3.81system 9:52.25elapsed 99%CPU (0avgtext+0avgdata 0maxresident)k0inputs+0outputs (0major+88206minor)pagefaults 0swaps-- statistics of differences bewteen nrlu and lu solver based on elevationtotal null and non-null cells: 131250 total null cells: 0 Of the non-null cells: ---------------------- n: 131250 minimum: 0 maximum: 0 range: 0 mean: 0 mean of absolute values: 0 standard deviation: 0 variance: 0 variation coefficient: nan % sum: 0 -- statistics of differences bewteen nrlu and lu solver based on slope total null and non-null cells: 131250 total null cells: 0 Of the non-null cells: ---------------------- n: 131250 minimum: 0 maximum: 0 range: 0 mean: 0 mean of absolute values: 0 standard deviation: 0 variance: 0 variation coefficient: nan % sum: 0 ########################################################## ++ Testing solver gauss with map elev_lidrural_mrpts578.09user 3.75system 9:42.95elapsed 99%CPU (0avgtext+0avgdata 0maxresident)k0inputs+0outputs (0major+87672minor)pagefaults 0swaps-- statistics of differences bewteen nrlu and gauss solver based on elevationtotal null and non-null cells: 131250 total null cells: 0 Of the non-null cells: ---------------------- n: 131250 minimum: 0 maximum: 0 range: 0 mean: 0 mean of absolute values: 0 standard deviation: 0 variance: 0 variation coefficient: nan % sum: 0-- statistics of differences bewteen nrlu and gauss solver based on slopetotal null and non-null cells: 131250 total null cells: 0 Of the non-null cells: ---------------------- n: 131250 minimum: 0 maximum: 0 range: 0 mean: 0 mean of absolute values: 0 standard deviation: 0 variance: 0 variation coefficient: nan % sum: 0 ########################################################## ++ Testing solver cg with map elev_lidrural_mrpts233.33user 1.98system 4:10.17elapsed 94%CPU (0avgtext+0avgdata 0maxresident)k0inputs+0outputs (0major+88477minor)pagefaults 0swaps-- statistics of differences bewteen nrlu and cg solver based on elevationtotal null and non-null cells: 131250 total null cells: 0 Of the non-null cells: ---------------------- n: 131250 minimum: -3.05176e-05 maximum: 3.05176e-05 range: 6.10352e-05 mean: -1.02307e-08 mean of absolute values: 5.68266e-07 standard deviation: 4.16437e-06 variance: 1.7342e-11 variation coefficient: -40704.9 % sum: -0.0013427734 -- statistics of differences bewteen nrlu and cg solver based on slope total null and non-null cells: 131250 total null cells: 0 Of the non-null cells: ---------------------- n: 131250 minimum: -3.05176e-05 maximum: 3.05176e-05 range: 6.10352e-05 mean: -1.02307e-08 mean of absolute values: 5.68266e-07 standard deviation: 4.16437e-06 variance: 1.7342e-11 variation coefficient: -40704.9 % sum: -0.0013427734 ########################################################## ++ Testing solver bicgstab with map elev_lidrural_mrpts359.87user 2.86system 6:03.24elapsed 99%CPU (0avgtext+0avgdata 0maxresident)k0inputs+0outputs (0major+89273minor)pagefaults 0swaps-- statistics of differences bewteen nrlu and bicgstab solver based on elevationtotal null and non-null cells: 131250 total null cells: 0 Of the non-null cells: ---------------------- n: 131250 minimum: -0.000244141 maximum: 0.000518799 range: 0.000762939 mean: -1.28813e-07 mean of absolute values: 6.98475e-07 standard deviation: 9.99815e-06 variance: 9.99631e-11 variation coefficient: -7761.74 % sum: -0.0169067383-- statistics of differences bewteen nrlu and bicgstab solver based on slopetotal null and non-null cells: 131250 total null cells: 0 Of the non-null cells: ---------------------- n: 131250 minimum: -0.000244141 maximum: 0.000518799 range: 0.000762939 mean: -1.28813e-07 mean of absolute values: 6.98475e-07 standard deviation: 9.99815e-06 variance: 9.99631e-11 variation coefficient: -7761.74 % sum: -0.0169067383 ########################################################## ++ Testing solver nrlu with map elev_contour_3m126.94user 0.88system 2:08.91elapsed 99%CPU (0avgtext+0avgdata 0maxresident)k0inputs+0outputs (0major+13106minor)pagefaults 0swaps ########################################################## ++ Testing solver lu with map elev_contour_3m157.57user 1.10system 2:38.99elapsed 99%CPU (0avgtext+0avgdata 0maxresident)k0inputs+0outputs (0major+13143minor)pagefaults 0swaps-- statistics of differences bewteen nrlu and lu solver based on elevationtotal null and non-null cells: 131250 total null cells: 0 Of the non-null cells: ---------------------- n: 131250 minimum: 0 maximum: 0 range: 0 mean: 0 mean of absolute values: 0 standard deviation: 0 variance: 0 variation coefficient: nan % sum: 0 -- statistics of differences bewteen nrlu and lu solver based on slope total null and non-null cells: 131250 total null cells: 0 Of the non-null cells: ---------------------- n: 131250 minimum: 0 maximum: 0 range: 0 mean: 0 mean of absolute values: 0 standard deviation: 0 variance: 0 variation coefficient: nan % sum: 0 ########################################################## ++ Testing solver gauss with map elev_contour_3m158.20user 1.08system 2:39.61elapsed 99%CPU (0avgtext+0avgdata 0maxresident)k0inputs+0outputs (0major+13074minor)pagefaults 0swaps-- statistics of differences bewteen nrlu and gauss solver based on elevationtotal null and non-null cells: 131250 total null cells: 0 Of the non-null cells: ---------------------- n: 131250 minimum: 0 maximum: 0 range: 0 mean: 0 mean of absolute values: 0 standard deviation: 0 variance: 0 variation coefficient: nan % sum: 0-- statistics of differences bewteen nrlu and gauss solver based on slopetotal null and non-null cells: 131250 total null cells: 0 Of the non-null cells: ---------------------- n: 131250 minimum: 0 maximum: 0 range: 0 mean: 0 mean of absolute values: 0 standard deviation: 0 variance: 0 variation coefficient: nan % sum: 0 ########################################################## ++ Testing solver cg with map elev_contour_3m69.81user 0.65system 1:10.61elapsed 99%CPU (0avgtext+0avgdata 0maxresident)k0inputs+0outputs (0major+13184minor)pagefaults 0swaps-- statistics of differences bewteen nrlu and cg solver based on elevationtotal null and non-null cells: 131250 total null cells: 0 Of the non-null cells: ---------------------- n: 131250 minimum: -1.52588e-05 maximum: 1.52588e-05 range: 3.05176e-05 mean: -8.77162e-08 mean of absolute values: 5.67859e-07 standard deviation: 2.28233e-06 variance: 5.20904e-12 variation coefficient: -2601.95 % sum: -0.0115127563 -- statistics of differences bewteen nrlu and cg solver based on slope total null and non-null cells: 131250 total null cells: 0 Of the non-null cells: ---------------------- n: 131250 minimum: -1.52588e-05 maximum: 1.52588e-05 range: 3.05176e-05 mean: -8.77162e-08 mean of absolute values: 5.67859e-07 standard deviation: 2.28233e-06 variance: 5.20904e-12 variation coefficient: -2601.95 % sum: -0.0115127563 ########################################################## ++ Testing solver bicgstab with map elev_contour_3m93.88user 0.72system 1:34.81elapsed 99%CPU (0avgtext+0avgdata 0maxresident)k0inputs+0outputs (0major+13286minor)pagefaults 0swaps-- statistics of differences bewteen nrlu and bicgstab solver based on elevationtotal null and non-null cells: 131250 total null cells: 0 Of the non-null cells: ---------------------- n: 131250 minimum: -3.8147e-05 maximum: 3.8147e-05 range: 7.62939e-05 mean: -2.70705e-07 mean of absolute values: 5.87391e-07 standard deviation: 3.21259e-06 variance: 1.03207e-11 variation coefficient: -1186.75 % sum: -0.0355300903-- statistics of differences bewteen nrlu and bicgstab solver based on slopetotal null and non-null cells: 131250 total null cells: 0 Of the non-null cells: ---------------------- n: 131250 minimum: -3.8147e-05 maximum: 3.8147e-05 range: 7.62939e-05 mean: -2.70705e-07 mean of absolute values: 5.87391e-07 standard deviation: 3.21259e-06 variance: 1.03207e-11 variation coefficient: -1186.75 % sum: -0.0355300903 ########################################################## ++ Testing solver nrlu with map elev_ned10m_cont10m57.10user 0.52system 0:57.80elapsed 99%CPU (0avgtext+0avgdata 0maxresident)k0inputs+0outputs (0major+22349minor)pagefaults 0swaps ########################################################## ++ Testing solver lu with map elev_ned10m_cont10m71.86user 0.62system 1:12.66elapsed 99%CPU (0avgtext+0avgdata 0maxresident)k0inputs+0outputs (0major+22374minor)pagefaults 0swaps-- statistics of differences bewteen nrlu and lu solver based on elevationtotal null and non-null cells: 131250 total null cells: 0 Of the non-null cells: ---------------------- n: 131250 minimum: 0 maximum: 0 range: 0 mean: 0 mean of absolute values: 0 standard deviation: 0 variance: 0 variation coefficient: nan % sum: 0 -- statistics of differences bewteen nrlu and lu solver based on slope total null and non-null cells: 131250 total null cells: 0 Of the non-null cells: ---------------------- n: 131250 minimum: 0 maximum: 0 range: 0 mean: 0 mean of absolute values: 0 standard deviation: 0 variance: 0 variation coefficient: nan % sum: 0 ########################################################## ++ Testing solver gauss with map elev_ned10m_cont10m71.32user 0.56system 1:12.09elapsed 99%CPU (0avgtext+0avgdata 0maxresident)k0inputs+0outputs (0major+22339minor)pagefaults 0swaps-- statistics of differences bewteen nrlu and gauss solver based on elevationtotal null and non-null cells: 131250 total null cells: 0 Of the non-null cells: ---------------------- n: 131250 minimum: 0 maximum: 0 range: 0 mean: 0 mean of absolute values: 0 standard deviation: 0 variance: 0 variation coefficient: nan % sum: 0-- statistics of differences bewteen nrlu and gauss solver based on slopetotal null and non-null cells: 131250 total null cells: 0 Of the non-null cells: ---------------------- n: 131250 minimum: 0 maximum: 0 range: 0 mean: 0 mean of absolute values: 0 standard deviation: 0 variance: 0 variation coefficient: nan % sum: 0 ########################################################## ++ Testing solver cg with map elev_ned10m_cont10m40.79user 0.38system 0:41.30elapsed 99%CPU (0avgtext+0avgdata 0maxresident)k0inputs+0outputs (0major+22403minor)pagefaults 0swaps-- statistics of differences bewteen nrlu and cg solver based on elevationtotal null and non-null cells: 131250 total null cells: 0 Of the non-null cells: ---------------------- n: 131250 minimum: -1.52588e-05 maximum: 1.52588e-05 range: 3.05176e-05 mean: 5.26356e-07 mean of absolute values: 1.0859e-06 standard deviation: 2.93473e-06 variance: 8.61265e-12 variation coefficient: 557.557 % sum: 0.0690841675 -- statistics of differences bewteen nrlu and cg solver based on slope total null and non-null cells: 131250 total null cells: 0 Of the non-null cells: ---------------------- n: 131250 minimum: -1.52588e-05 maximum: 1.52588e-05 range: 3.05176e-05 mean: 5.26356e-07 mean of absolute values: 1.0859e-06 standard deviation: 2.93473e-06 variance: 8.61265e-12 variation coefficient: 557.557 % sum: 0.0690841675 ########################################################## ++ Testing solver bicgstab with map elev_ned10m_cont10m50.83user 0.52system 0:51.49elapsed 99%CPU (0avgtext+0avgdata 0maxresident)k0inputs+0outputs (0major+22467minor)pagefaults 0swaps-- statistics of differences bewteen nrlu and bicgstab solver based on elevationtotal null and non-null cells: 131250 total null cells: 0 Of the non-null cells: ---------------------- n: 131250 minimum: -1.52588e-05 maximum: 1.52588e-05 range: 3.05176e-05 mean: -1.2271e-07 mean of absolute values: 4.69854e-07 standard deviation: 2.02924e-06 variance: 4.1178e-12 variation coefficient: -1653.69 % sum: -0.0161056519-- statistics of differences bewteen nrlu and bicgstab solver based on slopetotal null and non-null cells: 131250 total null cells: 0 Of the non-null cells: ---------------------- n: 131250 minimum: -1.52588e-05 maximum: 1.52588e-05 range: 3.05176e-05 mean: -1.2271e-07 mean of absolute values: 4.69854e-07 standard deviation: 2.02924e-06 variance: 4.1178e-12 variation coefficient: -1653.69 % sum: -0.0161056519
RST_solver_test_bar_charts.pdf
Description: Adobe PDF document
_______________________________________________ grass-dev mailing list [email protected] http://grass.itc.it/mailman/listinfo/grass-dev
