TY - JOUR
T1 - TopADDPi
T2 - An Affordable and Sustainable Raspberry Pi Cluster for Parallel-Computing Topology Optimization
AU - Zhang, Zhi Dong
AU - Yu, Dao Yuan
AU - Ibhadode, Osezua
AU - Meng, Liang
AU - Gao, Tong
AU - Zhu, Ji Hong
AU - Zhang, Wei Hong
N1 - Publisher Copyright:
© 2025 by the authors.
PY - 2025/3
Y1 - 2025/3
N2 - Parallel-Computing Topology Optimization (PCTO) has gained importance, especially with the advancement of additive manufacturing (AM), due to its ability to tackle high-dimensional, high-resolution challenges. PCTO is highly relevant to sustainable manufacturing processes and technologies, enabling resource-efficient designs, reduced emissions, and advancements in Industry 4.0 integration. However, PCTO poses difficulties for newcomers or researchers, mainly because of its reliance on non-traditional computing environments and the limited availability of high-performance computing (HPC) resources. Addressing this, the study introduces TopADDPi, a Raspberry Pi-based cluster system, which has been purpose-built to facilitate learning and research in PCTO. It provides detailed instructions for assembling and configuring a Raspberry Pi cluster, with a focus on cost-effectiveness and ease of use. The study thoroughly investigates how different hardware and software configurations affect computing efficiency. In addition, through extensive numerical testing, the performance, energy consumption, and environmental impact of the Raspberry Pi cluster are benchmarked against conventional computing systems. The findings demonstrate the cluster’s advantages in handling parallel computing, its indispensable role in debugging, its remarkable energy efficiency, and its significantly reduced carbon footprint compared to conventional systems. These attributes establish the Raspberry Pi cluster as an invaluable tool for both educational and research applications in structural engineering, offering an affordable, sustainable, and indispensable solution for PCTO.
AB - Parallel-Computing Topology Optimization (PCTO) has gained importance, especially with the advancement of additive manufacturing (AM), due to its ability to tackle high-dimensional, high-resolution challenges. PCTO is highly relevant to sustainable manufacturing processes and technologies, enabling resource-efficient designs, reduced emissions, and advancements in Industry 4.0 integration. However, PCTO poses difficulties for newcomers or researchers, mainly because of its reliance on non-traditional computing environments and the limited availability of high-performance computing (HPC) resources. Addressing this, the study introduces TopADDPi, a Raspberry Pi-based cluster system, which has been purpose-built to facilitate learning and research in PCTO. It provides detailed instructions for assembling and configuring a Raspberry Pi cluster, with a focus on cost-effectiveness and ease of use. The study thoroughly investigates how different hardware and software configurations affect computing efficiency. In addition, through extensive numerical testing, the performance, energy consumption, and environmental impact of the Raspberry Pi cluster are benchmarked against conventional computing systems. The findings demonstrate the cluster’s advantages in handling parallel computing, its indispensable role in debugging, its remarkable energy efficiency, and its significantly reduced carbon footprint compared to conventional systems. These attributes establish the Raspberry Pi cluster as an invaluable tool for both educational and research applications in structural engineering, offering an affordable, sustainable, and indispensable solution for PCTO.
KW - 3D printing
KW - arbitrary design domain
KW - parallel computing
KW - raspberry pi cluster
KW - sustainability
KW - topology optimization
UR - http://www.scopus.com/inward/record.url?scp=105001134105&partnerID=8YFLogxK
U2 - 10.3390/pr13030633
DO - 10.3390/pr13030633
M3 - 文章
AN - SCOPUS:105001134105
SN - 2227-9717
VL - 13
JO - Processes
JF - Processes
IS - 3
M1 - 633
ER -
