Intel® Parallel Computing Center at The University of Oklahoma

Principal Investigators:

Randall L. Kolar, David Ross Boyd and Austin Presidential Professor, Director of the School of Civil Engineering and Environmental Science
Kendra M. Dresback, Research Assistant Professor, School of Civil Engineering and Environmental Science

Description:

ADCIRC (Advanced CIRCulation) is a community-based shallow water model. ADCIRC model development has largely been driven by real-world, time critical applications.  Recent prominent examples include the following: (1) After Hurricane Katrina, Congress directed FEMA to use ADCIRC to develop their new flood inundation maps for US coastal areas;   (2) After the Fukushima disaster in Japan, the Nuclear Regulatory Commission is revising guidelines to use ADCIRC in flood studies for nuclear power plants; (3) NOAA, DHS, the Corps of Engineers and others use ADCIRC for prediction of storm surge and flooding; (4) The State of Louisiana uses ADCIRC as a planning tool for their coastal restoration plan; (5) The Corps of Engineers, an original ADCIRC sponsor and continued supporter, uses ADCIRC in many applications, including design enhancements to New Orleans’ hurricane protection system and updating a tidal database for the continental US; (6) After the Deepwater Horizon oil spill, a highly efficient particle tracking routine was coupled to the ADCIRC hydrodynamics to predict surface migration of the resulting oil slick; (7) A software superstructure, called the ADCIRC Surge Guidance System (ASGS), was developed for real-time applications by ingesting meteorological and hydrologic data and producing real-time predictions of flooding extents due to tropical and extra-tropical storms (see nc-cera link below).  

Much of this acceptance and use of ADCIRC is because it was one of the earliest unstructured mesh shallow water models to suppress spurious modes without resorting to artificial or numerical damping, owing to the superior wave propagation properties of the Generalized Wave Continuity Equation. A common characteristic of the aforementioned applications, and a strength of the model, is ADCIRC’s ability to support graded meshes using triangular finite elements, thus providing high resolution in areas of interest and coarser resolution elsewhere. For more than 25 years, ADCIRC has been continually enhanced in order to include more model physics and improve its performance. Current features include the following: 2D depth-averaged or 3D barotropic and baroclinic modes; fully implicit, semi-implicit, or explicit time marching schemes; parallel implementation; variable bottom-friction representation; sector-based and directionally-dependent wind drag coefficients; dynamic coupling to a wave energy equation model, SWAN; one-way coupling to upland hydrologic runoff models; and self-contained wind models (e.g., parametric and PBL). 

High-resolution, real-time simulations demand that the code be computationally efficient. For the current MPI version on parallel HPC architectures, ADCIRC utilizes the METIS software to partition the global problem into sub-domains, one per CPU core. Scaling studies show that ADCIRC is very efficient, with near optimal speedup for sufficiently large numbers of vertices per process. OU’s Intel® Parallel Computing Center focuses on ADCIRC optimization via OpenMP threading and vectorization, using Xeon Phi™ coprocessor natively as a host, as a mechanism for optimizing the code for both Xeon Phi™ coprocessors and Xeon CPUs.

Related websites:

http://adcirc.org/
http://nc-cera.renci.org

For more complete information about compiler optimizations, see our Optimization Notice.