The Sunway TaihuLight supercomputer is the world's first system with a peak performance greater than 100 PFlops. In this paper, we provide a detailed introduction to the TaihuLight system. In contrast with other existing heterogeneous supercomputers, which include both CPU processors and PCIe-connected many-core accelerators (NVIDIA GPU or Intel Xeon Phi), the computing power of TaihuLight is provided by a homegrown many-core SW26010 CPU that includes both the management processing elements (MPEs) and computing processing elements (CPEs) in one chip. With 260 processing elements in one CPU, a single SW26010 provides a peak performance of over three TFlops. To alleviate the memory bandwidth bottleneck in most applications, each CPE comes with a scratch pad memory, which serves as a user-controlled cache. To support the parallelization of programs on the new many-core architecture, in addition to the basic C/C++ and Fortran compilers, the system provides a customized Sunway OpenACC tool that supports the OpenACC 2.0 syntax. This paper also reports our preliminary efforts on developing and optimizing applications on the TaihuLight system, focusing on key application domains, such as earth system modeling, ocean surface wave modeling, atomistic simulation, and phase-field simulation.

The Beijing Climate Center atmospheric general circulation model version 2.0.1 (BCC_AGCM2.0.1) is described and its performance in simulating the present-day climate is assessed. BCC_AGCM2.0.1 originates from the community atmospheric model version 3 (CAM3) developed by the National Center for Atmospheric Research (NCAR). The dynamics in BCC_AGCM2.0.1 is, however, substantially different from the Eulerian spectral formulation of the dynamical equations in CAM3, and several new physical parameterizations have replaced the corresponding original ones. The major modification of the model physics in BCC_AGCM2.0.1 includes a new convection scheme, a dry adiabatic adjustment scheme in which potential temperature is conserved, a modified scheme to calculate the sensible heat and moisture fluxes over the open ocean which takes into account the effect of ocean waves on the latent and sensible heat fluxes, and an empirical equation to compute the snow cover fraction. Specially, the new convection scheme in BCC_AGCM2.0.1, which is generated from the Zhang and McFarlane's scheme but modified, is tested to have significant improvement in tropical maximum but also the subtropical minimum precipitation, and the modified scheme for turbulent fluxes are validated using EPIC2001 in situ observations and show a large improvement than its original scheme in CAM3. BCC_AGCM2.0.1 is forced by observed monthly varying sea surface temperatures and sea ice concentrations during 1949-2000. The model climatology is compiled for the period 1971-2000 and compared with the ERA-40 reanalysis products. The model performance is evaluated in terms of energy budgets, precipitation, sea level pressure, air temperature, geopotential height, and atmospheric circulation, as well as their seasonal variations. Results show that BCC_AGCM2.0.1 reproduces fairly well the present-day climate. The combined effect of the new dynamical core and the updated physical parameterizations in BCC_AGCM2.0.1 leads to an overall improvement, compared to the original CAM3.

Abstract. With semiconductor technology gradually approaching its physical and thermal limits, recent supercomputers have adopted major architectural changes to continue increasing the performance through more power-efficient heterogeneous many-core systems. Examples include Sunway TaihuLight that has four management processing elements (MPEs) and 256 computing processing elements (CPEs) inside one processor and Summit that has two central processing units (CPUs) and six graphics processing units (GPUs) inside one node. Meanwhile, current high-resolution Earth system models that desperately require more computing power generally consist of millions of lines of legacy code developed for traditional homogeneous multicore processors and cannot automatically benefit from the advancement of supercomputer hardware. As a result, refactoring and optimizing the legacy models for new architectures become key challenges along the road of taking advantage of greener and faster supercomputers, providing better support for the global climate research community and contributing to the long-lasting societal task of addressing long-term climate change. This article reports the efforts of a large group in the International Laboratory for High-Resolution Earth System Prediction (iHESP) that was established by the cooperation of Qingdao Pilot National Laboratory for Marine Science and Technology (QNLM), Texas A&M University (TAMU), and the National Center for Atmospheric Research (NCAR), with the goal of enabling highly efficient simulations of the high-resolution (25 km atmosphere and 10 km ocean) Community Earth System Model (CESM-HR) on Sunway TaihuLight. The refactoring and optimizing efforts have improved the simulation speed of CESM-HR from 1 SYPD (simulation years per day) to 3.4 SYPD (with output disabled) and supported several hundred years of pre-industrial control simulations. With further strategies on deeper refactoring and optimizing for remaining computing hotspots, as well as redesigning architecture-oriented algorithms, we expect an equivalent or even better efficiency to be gained on the new platform than traditional homogeneous CPU platforms. The refactoring and optimizing processes detailed in this paper on the Sunway system should have implications for similar efforts on other heterogeneous many-core systems such as GPU-based high-performance computing (HPC) systems.

No abstract