Scholarlink

Finite element methods have a crucial role in many engineering applications. With the utilization of high-performance computing, the scope of realizable models has expanded exponentially, while maintaining reasonable time to solution expectations. In the context of this study, we dive into the performance analysis of two distinct applications the (i) structural mechanics with the b2000++pro software, and (ii) mesh deformation with the FSMeshDeformation plugin from the FlowSimulator universe. Our investigation begins with the extraction of traces obtained from simulation runs using Score-P. Traces are subsequently subjected to comprehensive scrutiny using the visualization tool Vampir. To further refine our analysis, we strategically employ LIKWID to examine the most time-intensive segments of the code. Obtained results are then translated into the Roofline model. The investigation reveals that the performance of the analyzed hot loops is noticeably lower than the theoretical Roofline performance. This finding signifies that a noteworthy portion of the total runtime is neither limited by floating-point performance nor by available memory bandwidth. Options to close this performance gap are further discussed in this contribution.

错误：搜索内容不能为空，请输入英文关键词
错误：关键词超出字数限制，请精简
高级检索

Accelerating the FlowSimulator: Performance Analysis of Finite Element Methods on High–Performance Computers

  • 会议论文

  • 发表日期：

    2026年

查看全文
  • Neda Ebrahimi Pour,
  • Marco Cristofaro,
  • Immo Huismann,
  • Jana Gericke-Schuster,
  • Johannes Wendler

摘要

Finite element methods have a crucial role in many engineering applications. With the utilization of high-performance computing, the scope of realizable models has expanded exponentially, while maintaining reasonable time to solution expectations. In the context of this study, we dive into the performance analysis of two distinct applications the (i) structural mechanics with the b2000++pro software, and (ii) mesh deformation with the FSMeshDeformation plugin from the FlowSimulator universe. Our investigation begins with the extraction of traces obtained from simulation runs using Score-P. Traces are subsequently subjected to comprehensive scrutiny using the visualization tool Vampir. To further refine our analysis, we strategically employ LIKWID to examine the most time-intensive segments of the code. Obtained results are then translated into the Roofline model. The investigation reveals that the performance of the analyzed hot loops is noticeably lower than the theoretical Roofline performance. This finding signifies that a noteworthy portion of the total runtime is neither limited by floating-point performance nor by available memory bandwidth. Options to close this performance gap are further discussed in this contribution.