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.

Compensatory growth responses of Leymus chinensis, a dominant species in Inner Mongolia steppe, to clipping defoliation were evaluated in a pot-cultivated experiment under different nutrient (N and P) and water availability conditions. Leymus chinensis exhibited over-compensatory growth at the light and moderate clipping intensities (20% and 40% aerial mass removed) with a greater accumulated aboveground biomass, higher relative growth rate (RGR), more rhizomatic tillers and a stimulation of compensatory photosynthesis to the remnant leaves as compared with those of the unclipped plants. Intense clipping (80% aerial mass removed), which removed most of the aboveground tissues, greatly reduced the growth of aboveground biomass in comparison with that of the unclipped plants. Nitrogen addition only slightly improved the biomass production and RGR in light and moderately clipped plants, and it did not allow plants in the intense clipping condition to over-compensate. Phosphorus addition had no obvious influences on the growth and physiological responses to clipping defoliation. These results indicated that nutrient addition could not compensate for the negative effects of severe clipping on the defoliated grass. On the other hand, there were no distinct positive responses under water deficiency condition for L. chinensis at all clipping intensities with a significant reduction of aboveground and belowground biomass, lower RGR, fewer rhizomatic tillers, and a lower net photosynthetic rate than other wet treatments. Additionally, the chlorophyll contents of remnant leaves gradually increased with the increase of clipping intensities in each treatment. In conclusion, although L. chinensis could compensate for tissues removal by some morphological and physiological responses, intense clipping and drought can result in a significant decrease of biomass and growth rate, even under enriched nutrition conditions.

No abstract