Dynamic Load Balancing of Corbino-TBG Electronic Transport Simulations on Heterogeneous Clusters
Resumo
Simulações de transporte eletrônico em grafeno bicamada torcido (TBG) na geometria Corbino resolvem grandes sistemas lineares esparsos a cada ponto de energia. Este trabalho atribui pontos de energia independentes sob demanda a trabalhadores CPU e GPU com solvers esparsos distintos, sem pesos de desempenho predefinidos. Para uma instância de 1,5◦, oito nós CPU atingem speedup 6,62 sobre um nó; dois nós CPU com os dois GK210 de uma Tesla K80 atingem 3,17 sobre a referência C/PARDISO, com 99,85% de eficiência de balanceamento e makespan 13,2% menor que a partição estática estimada. Os ramos diferem em no máximo 1,0×10−10 na condutância relativa.
Referências
AMReX-Microelectronics (2024). Eleqtronex: An open-source, first-principles, and full-band simulator for quantum-transport in electron-phonon-coupled nano-devices. Acessado em: 2025-06-19.
Assis, I. A. S., Oliveira, A. D. S., Barros, T., Sardina, I. M., Bianchini, C. P., and Xavier De-Souza, S. (2019). Distributed-memory load balancing with cyclic token-based work-stealing applied to reverse time migration. IEEE Access, 7:128419–128430.
Bahamon, D. A., Castro Neto, A. H., and Pereira, V. M. (2013). Effective contact model for geometry-independent conductance calculations in graphene. Physical Review B, 88(23):235433.
Ben-Nun, T., Schneider, T., Ziogas, A. N., Müller, T., Luisier, M., and Hoefler, T. (2019). Stateful dataflow multigraphs: A data-centric model for performance portability on heterogeneous architectures. In Proceedings of the International Conference for High Performance Computing, Networking, Storage and Analysis, SC ’19, New York, NY, USA. ACM. Gordon Bell Prize recipient.
Calderara, M., Brück, S., Pedersen, A., Bani-Hashemian, M. H., VandeVondele, J., and Luisier, M. (2015). Pushing back the limit of ab-initio quantum transport simulations on hybrid supercomputers. In Proceedings of the International Conference for High Performance Computing, Networking, Storage and Analysis, SC ’15, New York, NY, USA. ACM.
Datta, S. (1997). Electronic transport in mesoscopic systems. Cambridge University Press, Cambridge.
Datta, S. (2005). Quantum transport: atom to transistor. Cambridge University Press, Cambridge.
Dongarra, J. J., Du Croz, J., Hammarling, S., and Duff, I. S. (1990). A Set of Level 3 Basic Linear Algebra Subprograms. ACM Transactions on Mathematical Software, 16(1):1–17.
Fernandes, J. B. (2020). Granularity load scheduling auto-tuning for multi-core processors applied to reverse-time migration. Dissertação (mestrado em engenharia elétrica e de computação), Universidade Federal do Rio Grande do Norte, Natal.
Ferry, D. and Goodnick, S. M. (1999). Transport in nanostructures. Cambridge University Press, Cambridge. No. 6.
Groth, C. W., Wimmer, M., Akhmerov, A. R., and Waintal, X. (2014). Kwant: a software package for quantum transport. New Journal of Physics, 16:063065.
Hoefler, T. and Belli, R. (2015). Scientific benchmarking of parallel computing systems: twelve ways to tell the masses when reporting performance results. In Proceedings of the International Conference for High Performance Computing, Networking, Storage and Analysis (SC ’15), pages 1–12. ACM.
Hénon, P., Ramet, P., and Roman, J. (2002). PaStiX: A high-performance parallel direct solver for large sparse linear systems. Parallel Computing, 28(2):301–321.
Inria Bordeaux – Sud-Ouest (2024). PaStiX solver documentation.
Intel Corporation (2023). Intel® oneapi math kernel library (onemkl) documentation.
Löthman, T., Schmidt, J., Parhizgar, F., Black-Schaffer, A. M., and Lado, J. L. (2022). Nematic superconductivity in magic-angle twisted bilayer graphene from atomistic modeling. Communications Physics, 5:92.
NanoTCAD ViDES (2023a). MPIze - NanoTCAD ViDES documentation.
NanoTCAD ViDES (2023b). NanoTCAD ViDES. [link]. Accessed: 2025-06-12.
Novoselov, K. S., Fal’ko, V. I., Colombo, L., Gellert, P. R., Schwab, M. G., and Kim, K. (2012). A roadmap for graphene. Nature, 490(7419):192–200.
NVIDIA Corporation (2026a). cuSOLVER library documentation.
NVIDIA Corporation (2026b). NVIDIA cuDSS: A high-performance CUDA library for direct sparse solvers — getting started.
Papior, N. R., Gunst, T., Stradi, D., and Stokbro, K. (2017). The TranSIESTA method for efficient and parallel electronic transport calculations. Computer Physics Communications, 212:8–24.
Passos, A. and Bianchini, C. (2025). Comparação de desempenho de implementações fortran e c: Corbino tbg. In Anais da XVI Escola Regional de Alto Desempenho de São Paulo, pages 98–101, Porto Alegre, RS, Brasil. SBC.
Pellegrini, F. and Roman, J. (1996). SCOTCH: A software package for static mapping by dual recursive bipartitioning of process and architecture graphs. In High-Performance Computing and Networking (HPCN Europe 1996), volume 1067 of Lecture Notes in Computer Science, pages 493–498. Springer.
Santana, C. d. S. (2020). Workload scheduling analysis in geophysical numerical methods. Dissertação (mestrado em engenharia elétrica e de computação), Universidade Federal do Rio Grande do Norte, Natal.
Sawant, S. S., Léonard, F., Yao, Z., and Nonaka, A. (2025). ELEQTRONeX: A GPU-accelerated exascale framework for non-equilibrium quantum transport in nanomaterials. npj Computational Materials, 11:110.
Scott, J. and Tůma, M. (2023). Algorithms for sparse linear systems. Springer, Cham.
Synopsys (2024). QuantumATK. [link]. Accessed: 2025-06-12.
Synopsys (2025a). Parallelization of QuantumATK calculations - QuantumATK documentation.
Synopsys (2025b). QuantumATK new release notes.
Synopsys (2025c). Technical hardware requirements - quantumatk documentation. [link]. Acessado em: 2025-06-14.
The SIESTA project (2024). TranSIESTA Benchmarks. [link]. Accessed: 2025-06-12.
Tomov, S., Dongarra, J., and Baboulin, M. (2010). Towards dense linear algebra for hybrid GPU accelerated manycore systems. Parallel Computing, 36(5–6):232–240.
Assis, I. A. S., Oliveira, A. D. S., Barros, T., Sardina, I. M., Bianchini, C. P., and Xavier De-Souza, S. (2019). Distributed-memory load balancing with cyclic token-based work-stealing applied to reverse time migration. IEEE Access, 7:128419–128430.
Bahamon, D. A., Castro Neto, A. H., and Pereira, V. M. (2013). Effective contact model for geometry-independent conductance calculations in graphene. Physical Review B, 88(23):235433.
Ben-Nun, T., Schneider, T., Ziogas, A. N., Müller, T., Luisier, M., and Hoefler, T. (2019). Stateful dataflow multigraphs: A data-centric model for performance portability on heterogeneous architectures. In Proceedings of the International Conference for High Performance Computing, Networking, Storage and Analysis, SC ’19, New York, NY, USA. ACM. Gordon Bell Prize recipient.
Calderara, M., Brück, S., Pedersen, A., Bani-Hashemian, M. H., VandeVondele, J., and Luisier, M. (2015). Pushing back the limit of ab-initio quantum transport simulations on hybrid supercomputers. In Proceedings of the International Conference for High Performance Computing, Networking, Storage and Analysis, SC ’15, New York, NY, USA. ACM.
Datta, S. (1997). Electronic transport in mesoscopic systems. Cambridge University Press, Cambridge.
Datta, S. (2005). Quantum transport: atom to transistor. Cambridge University Press, Cambridge.
Dongarra, J. J., Du Croz, J., Hammarling, S., and Duff, I. S. (1990). A Set of Level 3 Basic Linear Algebra Subprograms. ACM Transactions on Mathematical Software, 16(1):1–17.
Fernandes, J. B. (2020). Granularity load scheduling auto-tuning for multi-core processors applied to reverse-time migration. Dissertação (mestrado em engenharia elétrica e de computação), Universidade Federal do Rio Grande do Norte, Natal.
Ferry, D. and Goodnick, S. M. (1999). Transport in nanostructures. Cambridge University Press, Cambridge. No. 6.
Groth, C. W., Wimmer, M., Akhmerov, A. R., and Waintal, X. (2014). Kwant: a software package for quantum transport. New Journal of Physics, 16:063065.
Hoefler, T. and Belli, R. (2015). Scientific benchmarking of parallel computing systems: twelve ways to tell the masses when reporting performance results. In Proceedings of the International Conference for High Performance Computing, Networking, Storage and Analysis (SC ’15), pages 1–12. ACM.
Hénon, P., Ramet, P., and Roman, J. (2002). PaStiX: A high-performance parallel direct solver for large sparse linear systems. Parallel Computing, 28(2):301–321.
Inria Bordeaux – Sud-Ouest (2024). PaStiX solver documentation.
Intel Corporation (2023). Intel® oneapi math kernel library (onemkl) documentation.
Löthman, T., Schmidt, J., Parhizgar, F., Black-Schaffer, A. M., and Lado, J. L. (2022). Nematic superconductivity in magic-angle twisted bilayer graphene from atomistic modeling. Communications Physics, 5:92.
NanoTCAD ViDES (2023a). MPIze - NanoTCAD ViDES documentation.
NanoTCAD ViDES (2023b). NanoTCAD ViDES. [link]. Accessed: 2025-06-12.
Novoselov, K. S., Fal’ko, V. I., Colombo, L., Gellert, P. R., Schwab, M. G., and Kim, K. (2012). A roadmap for graphene. Nature, 490(7419):192–200.
NVIDIA Corporation (2026a). cuSOLVER library documentation.
NVIDIA Corporation (2026b). NVIDIA cuDSS: A high-performance CUDA library for direct sparse solvers — getting started.
Papior, N. R., Gunst, T., Stradi, D., and Stokbro, K. (2017). The TranSIESTA method for efficient and parallel electronic transport calculations. Computer Physics Communications, 212:8–24.
Passos, A. and Bianchini, C. (2025). Comparação de desempenho de implementações fortran e c: Corbino tbg. In Anais da XVI Escola Regional de Alto Desempenho de São Paulo, pages 98–101, Porto Alegre, RS, Brasil. SBC.
Pellegrini, F. and Roman, J. (1996). SCOTCH: A software package for static mapping by dual recursive bipartitioning of process and architecture graphs. In High-Performance Computing and Networking (HPCN Europe 1996), volume 1067 of Lecture Notes in Computer Science, pages 493–498. Springer.
Santana, C. d. S. (2020). Workload scheduling analysis in geophysical numerical methods. Dissertação (mestrado em engenharia elétrica e de computação), Universidade Federal do Rio Grande do Norte, Natal.
Sawant, S. S., Léonard, F., Yao, Z., and Nonaka, A. (2025). ELEQTRONeX: A GPU-accelerated exascale framework for non-equilibrium quantum transport in nanomaterials. npj Computational Materials, 11:110.
Scott, J. and Tůma, M. (2023). Algorithms for sparse linear systems. Springer, Cham.
Synopsys (2024). QuantumATK. [link]. Accessed: 2025-06-12.
Synopsys (2025a). Parallelization of QuantumATK calculations - QuantumATK documentation.
Synopsys (2025b). QuantumATK new release notes.
Synopsys (2025c). Technical hardware requirements - quantumatk documentation. [link]. Acessado em: 2025-06-14.
The SIESTA project (2024). TranSIESTA Benchmarks. [link]. Accessed: 2025-06-12.
Tomov, S., Dongarra, J., and Baboulin, M. (2010). Towards dense linear algebra for hybrid GPU accelerated manycore systems. Parallel Computing, 36(5–6):232–240.
Publicado
03/11/2026
Como Citar
PASSOS, Arthur M.; BIANCHINI, Calebe P..
Dynamic Load Balancing of Corbino-TBG Electronic Transport Simulations on Heterogeneous Clusters. In: SIMPÓSIO EM SISTEMAS COMPUTACIONAIS DE ALTO DESEMPENHO (SSCAD), 27. , 2026, Natal/RN.
Anais [...].
Porto Alegre: Sociedade Brasileira de Computação,
2026
.
p. 552-566.
DOI: https://doi.org/10.5753/sscad.2026.34984.
