Abstract
This paper proposes a well-suited strategy for High Performance Computing (HPC) of density-based topology optimization using Graphics Processing Units (GPUs). Such a strategy takes advantage of Massively Parallel Processing (MPP) architectures to overcome the computationally demanding procedures of density-based topology design, both in terms of memory consumption and processing time. This is done exploiting data locality and minimizing both memory consumption and data transfers. The proposed GPU instance makes use of different granularities for the topology optimization pipeline, which are selected to properly balance the workload between the threads exploiting the parallelization potential of massively parallel architectures. The performance of the fine-grained GPU instance of the solving stage is evaluated using two preconditioning techniques. The proposal is also compared with the classical CPU implementation for diverse topology optimization problems, including stiffness maximization, heat sink design and compliant mechanism design.
Highlights
Topology optimization aims to find the optimal distribution of material within a design domain such that an objective function is minimized under certain constraints [1]
This paper proposes a multi-granular Graphics Processing Units (GPUs) implementation of the different stages involved in density-based topology optimization methods
This paper has investigated about the proper strategy and techniques to achieve efficient calculation and reasonable speedups using GPU computing for the computationally intensive tasks of density-based topology optimization methods
Summary
Topology optimization aims to find the optimal distribution of material within a design domain such that an objective function is minimized under certain constraints [1]. The GPU instance of PCG solver using geometric multigrid preconditioning in topology optimization configured to perform a reduced number of FEAs and iterations per FEA, permitted Wu et al [37] to solve large-scale problems in a short time This is done configuring the iterative method with low tolerance level along with SIMP method using standard Optimality Criteria (OC) method [1]. The calculation of sensitivities, filter and density update are implemented using GPU computing in order not to limit the theoretical speedup according to Amdahl’s law [41] The granularity of such tasks is at the finite element level due to the nature of the operations do not allow us to reduce the grain size, which usually improves the GPU performance.
Sign-in/Register to view highlights & summary Sign-in/Register to access full text options 
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a callDisclaimer: All third-party content on this website/platform is and will remain the property of their respective owners and is provided on "as is" basis without any warranties, express or implied. Use of third-party content does not indicate any affiliation, sponsorship with or endorsement by them. Any references to third-party content is to identify the corresponding services and shall be considered fair use under The CopyrightLaw.
