lucarovai2
New member
Dear All,
I am working on the assimilation of data in the WRF-ARW 4.1 model. I need to produce a background error on the whole Mediterranean area with a horizontal resolution of 5 km and 40 vertical levels. I had no problems when I made it for a limited area, particularly in Italy. Although now my center has a lot of computing power, the procedure (a NMC method) stops (especially at stage 2: gen_be_stage2.e). This seems to be due to the fact that the program does not foresee a parallelization of the calculation and therefore, having to diagonalise a very large matrix (600,000 x 600,000 for 40 vertical levels) it fails to move forward. Do you have any suggestions?
L.
I am working on the assimilation of data in the WRF-ARW 4.1 model. I need to produce a background error on the whole Mediterranean area with a horizontal resolution of 5 km and 40 vertical levels. I had no problems when I made it for a limited area, particularly in Italy. Although now my center has a lot of computing power, the procedure (a NMC method) stops (especially at stage 2: gen_be_stage2.e). This seems to be due to the fact that the program does not foresee a parallelization of the calculation and therefore, having to diagonalise a very large matrix (600,000 x 600,000 for 40 vertical levels) it fails to move forward. Do you have any suggestions?
L.