Contributing author: John Dennis
While the emergence of many-core technology from Intel and Nvidia has illustrated great potential, capitalizing on this potential presents considerable challenges for large scientific applications. In particular we focus on the domain of climate modeling. Climate models typically have very large code bases, > 1 million lines of code which makes support of multiple version of the code infeasible. Climate models are computationally very expensive which places a premium on model performance. Furthermore because of the large user base of these applications, there is a need to support a wide range of computational platforms. These three characteristics of climate modeling make it a particularly challenging application domain for which to apply accelerator technology.
We describe work to migrate important climate applications used at the National Center for Atmospheric Research (NCAR) to Intel Xeon Phi and Nvidia GPU. We also provide a comparison with more conventional Intel SandyBridge with Infiniband. In particular we concentrate on both kernels from the High Order Methods Model Environment (HOMME) as well as the complete application. For the kernels we explored several different programming paradigms, including OpenACC, Cuda C/Fortran, and OpenMP. We will illustrate our use of Extrae a tracing tool developed at the Barcelona Supercomputing Center and the profiling tool TAU to isolate sections of code for further improvement. We will also present initial evaluation of the Community Earth System Model in different executions modes on the Intel Xeon Phi system at TACC.
Srinath Vadlamani obtained his Ph.D. from CU-Boulder in 2005 in Applied Mathematics with advisor Jim Meiss. Dr. Vadlamani did his post-doc for a year at the Univ. of Washington with Uri Shumlack developing finite volume methods for MHD simulations of innovative fusion confinement concepts. He returned to Boulder in 2006 and Tech-X Corporation working with algorithm development for MHD and gyrokinetic codes. He was part of the FACETS integrated modeling SCIDAC effort and created the interlagnuage interoperable generator that exposed fortran transport codes to the C++ FACETS framework. In Aug. 2012, Dr. Vadlamani joined NCAR's ASAP group to help with efficient usage of accelerated hardware for NCAR's climate codes.
After an undergraduate degree in Electronic Engineering at Dankook University of South Korea, Youngsung Kim has worked in mobile telecommunication industry for 13 years mostly as a software developer. In 2010, he returned back to school at University of Utah and majored in Scientific Computing. During the study, he participated WRF climate simulation project and brain image matching project along with taking core courses including numerical methods and parallel computing. After graduation with master's degree from Univ. of Utah, he joined NCAR and has been working on accelerator technologies until now.
If you use a non-flash enabled device, you may download the video here