Enhancing Supercomputing Education through a Low-Cost Cluster: A Case Study at Insper
DOI:
https://doi.org/10.5753/ijcae.2023.4824Abstract
High-Performance Computing (HPC) and parallel programming presents intricate challenges due to the sophisticated interplay between advanced hardware and software components. This paper delineates a case study of a cost-effective cluster comprising 24 Upboards engineered to bolster a project-based Supercomputing course. The project received the name of UpCluster, and it serves as a pragmatic, cost-efficient solution for experiential learning, mitigating the abstraction often associated with theoretical constructs. The curriculum encompasses various topics, including distributed computing, parallel computing, algorithm analysis, and the Message Passing Interface (MPI). The team meticulously documented the cluster infrastructure, providing a comprehensive guide for the configuration and utilization of the Single Board Computer cluster with Kubernetes and MPI operators. Students engaged in practical experimentation, developing scalable algorithms, and gaining valuable insights into the challenges and opportunities associated with distributed computing. These experiences fostered a deeper appreciation for the complexities and potential of distributed computing. The primary objective of this study is to demonstrate the efficacy of the cost-effective cluster in augmenting high-performance computing education. By providing a practical learning environment, the UpCluster complements theoretical instruction and empowers students to acquire practical skills in the design of large-scale distributed systems with multi-core nodes. Furthermore, the paper discusses this low-cost cluster’s potential impact and applications in HPC education. The insights from the study may benefit academic departments and institutions seeking to develop analogous project-based courses focused on high-performance computing for graduate students.
Downloads
Referências
J. Nengzhi, Z. Jianwu, X. Haili, W. Xiaoning, and S. Yulin, “Comparative research on high-speed networks of high performance computing cluster based on mpigraph,” pp. 579–583, 2020.
J. A. Shamsi, N. M. Durrani, and N. Kafi, “Novelties in teaching high performance computing,” pp. 772–778, 2015.
C. Weems, “Eduhipc 2021 keynote talk early parallel and distributed computing education: Canary in the coal mine,” pp. 3–3, 2021.
S. Diesburg, P. Gray, and D. Joiner, “High performance computing environments without the fuss: the bootable cluster cd,” pp. 8 pp.–, 2005.
V. W. Freeh, F. Pan, N. Kappiah, D. K. Lowenthal, and R. Springer, “Exploring the energy-time tradeoff in mpi programs on a power-scalable cluster,” 2005.
N. Regola and J. C. Ducom, “Recommendations for virtualization technologies in high performance computing,” 2010, pp. 409–416.
W. Mao, K. Li, Q. Cheng, L. Dai, B. Li, X. Xie, H. Li, L. Lin, and H. Yu, “A configurable floating-point multiple-precision processing element for hpc and ai converged computing,” IEEE Transactions on Very Large Scale Integration (VLSI) Systems, vol. 30, pp. 213–226, 2 2022.
C.-C. Chen, K. S. Khorassani, G. K. R. Kuncham, R. Vaidya, M. Abduljabbar, A. Shafi, H. Subramoni, and D. K. Panda, “Implementing and optimizing a gpu-aware mpi library for intel gpus: Early experiences.” IEEE, 5 2023, pp. 131–140. [Online]. Available: [link]
M. S. Beni, L. Crisci, and B. Cosenza, “Empi: Enhanced message passing interface in modern c++.” IEEE, 5 2023, pp. 141–153. [Online]. Available: [link]
N. Zhou, H. Zhou, and D. Hoppe, “Containerization for high performance computing systems: Survey and prospects,” IEEE Transactions on Software Engineering, vol. 49, pp. 2722–2740, 4 2023.
P. Gschwandtner, A. Hirsch, P. Thoman, P. Zangerl, H. Jordan, and T. Fahringer, “The cluster coffer: Teaching hpc on the road,” Journal of Parallel and Distributed Computing, vol. 155, pp. 50–62, 9 2021.
J. Mullen, L. Milechin, and D. Milechin, “Teaching and learning hpc through serious games,” Journal of Parallel and Distributed Computing, vol. 158, pp. 115–125, 12 2021.
S. Catalán, R. Carratalá-Sáez, and S. Iserte, “Leveraging teaching on demand: Approaching hpc to undergrads,” Journal of Parallel and Distributed Computing, vol. 156, pp. 148–162, 10 2021.
D. Huang, Q. Liu, J. Choi, N. Podhorszki, S. Klasky, J. Logan, G. Ostrouchov, X. He, and M. Wolf, “Can i/o variability be reduced on qos-less hpc storage systems?” IEEE Transactions on Computers, vol. 68, pp. 631–645, 5 2019.
D. Li, B. R. de Supinski, M. Schulz, K. Cameron, and D. S. Nikolopoulos, “Hybrid mpi/openmp power-aware computing,” pp. 1–12, 2010.
A. N. Jing Hui and B. S. Lee, “Epsilon: A microservices based distributed scheduler for kubernetes cluster,” pp. 1–6, 2021.
O. Mart, C. Negru, F. Pop, and A. Castiglione, “Observability in kubernetes cluster: Automatic anomalies detection using prometheus.” Institute of Electrical and Electronics Engineers Inc., 12 2020, pp. 565– 570.
B. T. Hasan and D. B. Abdullah, “Real-time resource monitoring framework in a heterogeneous kubernetes cluster.” Institute of Electrical and Electronics Engineers Inc., 2022, pp. 184–189.
B. Neelima, “High performance computing education in an indian engineering institute,” Journal of Parallel and Distributed Computing, vol. 105, pp. 73–82, 7 2017.
R. Rabenseifner, G. Hager, and G. Jost, “Hybrid mpi/openmp parallel programming on clusters of multi-core smp nodes parallel, distributed and network-based processing,” 2009.
B. Chaudhury, A. Varma, Y. Keswani, Y. Bhatnagar, and S. Parikh, “Let’s hpc: A web-based platform to aid parallel, distributed and high performance computing education,” Journal of Parallel and Distributed Computing, vol. 118, pp. 213–232, 8 2018.
Downloads
Published
Como Citar
Issue
Section
Licença
Copyright (c) 2023 Os autores
Este trabalho está licenciado sob uma licença Creative Commons Attribution-NonCommercial 4.0 International License.
