Today I check the compilation and running of tRIBS model preliminarily. The results are following:
I compile the code using g++-7.5, openmpi-4.1 on Ubuntu-20.04 OS. And I run the big_spring case using serial and parallelized tRIBS model without any errors. But I use pytribs to build the Newman_Canyon case, I meet some problems:
(1) I run the serial tRIBS, there's no errors. But in the running process, the screen printout is:
"No Case Identified in tHydroModel::UnSaturatedZone"
Meanwhile, in the step "Part 8: Hydrological Simulation loop", there's no information output.
(2) I run the parallelized tRIBS, the error happend as following:
"
Segmentation fault (11) with (signal code)
Address not mapped (1)
"
I check this problem in Google, someone told this problem was related with MPI. I have tested using openmpi-4.1 and mpich-3.3, the same error still happened.
(3) I compare the input files of big_spring and Newman_Canyon, I found that:
there are no DEMFILE and BedROCKFILE in Newman_Canyon case.
Wish feedback from you, and Thanks for your development.
I don't read the code of tRIBS. I just study the PIHM and SHUD hydrological model using openMP.
Yours,
Li Jian
Today I check the compilation and running of tRIBS model preliminarily. The results are following:
I compile the code using g++-7.5, openmpi-4.1 on Ubuntu-20.04 OS. And I run the big_spring case using serial and parallelized tRIBS model without any errors. But I use pytribs to build the Newman_Canyon case, I meet some problems:
(1) I run the serial tRIBS, there's no errors. But in the running process, the screen printout is:
"No Case Identified in tHydroModel::UnSaturatedZone"
Meanwhile, in the step "Part 8: Hydrological Simulation loop", there's no information output.
(2) I run the parallelized tRIBS, the error happend as following:
"
Segmentation fault (11) with (signal code)
Address not mapped (1)
"
I check this problem in Google, someone told this problem was related with MPI. I have tested using openmpi-4.1 and mpich-3.3, the same error still happened.
(3) I compare the input files of big_spring and Newman_Canyon, I found that:
there are no DEMFILE and BedROCKFILE in Newman_Canyon case.
Wish feedback from you, and Thanks for your development.
I don't read the code of tRIBS. I just study the PIHM and SHUD hydrological model using openMP.
Yours,
Li Jian