Task 213

Applications, Evaluation, and Improvement of Goddard Multi-scale Modeling System

Principal Investigator(s):

B-W. Shen


W. Tao

Last Updated:

October 26, 2012 15:25:47

Description of Problem

To examine the impacts of resolved moist processes on the short-term and long-term climate simulations with the MMF.

Scientific Objectives and Approach

With the new version of MMF from Jiundar Chern in June, 2010, I started re-implementing the revised parallelism into the MMF to improve its scalability for the feasibility of performing long-term simulations. In addition, I worked with Bryan of ARC to port and integrate software modules for fusing satellite data such as TRMM and QuikSCAT for inter-comparisons with model results. To make the new version of MMF scalable, we first integrated thousand copies of the GCEs into a super-component called meta-global GCE (mgGCE), implemented a revised parallelism with a 2D domain decomposition, and then coupled the mgGCE and fvGCM into the new version of MMF (Related tasks include the test and changes in the compiler flags with the new Fortran compiler and new modules).


• Published 3 referred journal articles (AGU JGR, AGU GRL and IEEE CiSE)

• Jointly developed data modules with Bryan of NASA/ARC to fuse NASA satellite data such as QuikSCAT sea winds and TRMM precipitation for inter-comparisons with model simulations (Shen et al., 2010c).

• Re-implemented a revised parallelism in the new version of MMF by integrating the massive (12,960) copies of GCEs into a super-component called “meta-global Goddard Ensemble Model (mg-GCE)” and developing a coupler between the mgGCE and the fvGCM for data regridding (e.g., Shen et al., 2011, a AIST annual report, to be submitted to the NASA ESTF 2011).

• Conducted preliminary benchmarks with the MMF v2.0, showing that this version is scalable up to 364 CPUs, while the earlier version (v1.0) is only scalable to 30 CPUs. A speedup of 12.2 is obtained by increasing the number of CPUs from 30 to 364. Further improvement is being planned and conducted with the support of the application group at NASA/ARC (Samson).

Refereed Journal Publications

Shen, B.-W., W.-K. Tao, and B. Green, 2010c: Coupling NASA Advanced Multi-Scale Modeling and Concurrent Visualization Systems for Improving Predictions of Tropical High-Impact Weather (CAMVis). Computing in Science and Engineering, 23 Nov. 2010. IEEE computer Society Digital Library. IEEE Computer Society, http://doi.ieeecomputersociety.org/10.1109/MCSE.2010.141 (impact factor 0.973)

Shen, B.-W., W.-K. Tao, and M.-L. C. Wu, 2010b: African Easterly Waves in 30-day High-resolution Global Simulations: A Case Study during the 2006 NAMMA Period. Geophys. Res. Lett., L18803, doi:10.1029/2010GL044355. (impact factor 3.204) http://atmospheres.gsfc.nasa.gov/cloud_modeling/docs/2010GL044355.pdf

Shen, B.-W., W.-K. Tao, W. K. Lau, R. Atlas, 2010a: Predicting Tropical Cyclogenesis with a Global Mesoscale Model: Hierarchical Multiscale Interactions During the Formation of Tropical Cyclone Nargis (2008) . J. Geophys. Res., 115, D14102, doi:10.1029/2009JD013140. (impact factor 3.082) http://atmospheres.gsfc.nasa.gov/cloud_modeling/docs/Shen_2009JD013140.pdf

Other Publications and Conferences