Browsing by Author "Meng, Jie"

Now showing 1 - 2 of 2
  • Thumbnail Image
    Item
    Accelerating High-Order Stencils on GPUs
    (IEEE, 2020) Sai, Ryuichi; Mellor-Crummey, John; Meng, Xiaozhu; Araya-Polo, Mauricio; Meng, Jie
    While implementation strategies for low-order stencils on GPUs have been well-studied in the literature, not all of the techniques work well for high-order stencils, such as those used for seismic imaging. In this paper, we study practical seismic imaging computations on GPUs using high-order stencils on large domains with meaningful boundary conditions. We manually crafted a collection of implementations of a 25-point seismic modeling stencil in CUDA along with code to apply the boundary conditions. We evaluated our stencil code shapes, memory hierarchy usage, data-fetching patterns, and other performance attributes. We conducted an empirical evaluation of these stencils using several mature and emerging tools and discuss our quantitative findings. Some of our implementations achieved twice the performance of a proprietary code developed in C and mapped to GPUs using OpenACC. Additionally, several of our implementations have excellent performance portability.
  • Thumbnail Image
    Item
    Phonon-Assisted Hot Carrier Generation in Plasmonic Semiconductor Systems
    (American Chemical Society, 2021) Hattori, Yocefu; Meng, Jie; Zheng, Kaibo; Meier de Andrade, Ageo; Kullgren, Jolla; Broqvist, Peter; Nordlander, Peter; Sá, Jacinto
    Plasmonic materials have optical cross sections that exceed by 10-fold their geometric sizes, making them uniquely suitable to convert light into electrical charges. Harvesting plasmon-generated hot carriers is of interest for the broad fields of photovoltaics and photocatalysis; however, their direct utilization is limited by their ultrafast thermalization in metals. To prolong the lifetime of hot carriers, one can place acceptor materials, such as semiconductors, in direct contact with the plasmonic system. Herein, we report the effect of operating temperature on hot electron generation and transfer to a suitable semiconductor. We found that an increase in the operation temperature improves hot electron harvesting in a plasmonic semiconductor hybrid system, contrasting what is observed on photodriven processes in nonplasmonic systems. The effect appears to be related to an enhancement in hot carrier generation due to phonon coupling. This discovery provides a new strategy for optimization of photodriven energy production and chemical synthesis.