Exploração de espaço de projeto para multicores heterogêneos com o uso aprendizado de máquina: o estado da arte / Using machine learning to perform design space exploration of heterogeneous multicore architectures: a state of art review

Alba Sandyra Bezerra Lopes, Monica Magalhães Pereira

Abstract


A cada ano aumenta-se a demanda por recursos computacionais das aplicações que executam em sistemas embarcados. Para atender a essa demanda, os projetos desses sistemas têm feito uso da combinação de componentes diversificados, resultando em plataformas heterogêneas que buscam balancear o poder de processamento com o consumo de energia. Entretanto, uma questão chave no projeto desses sistemas é quais componentes combinar para atender ao desempenho esperado ao custo de área e energia adicionais. Realizar uma vasta exploração do espaço de projeto (DSE) permite mensurar previamente custo dessas plataformas antes da fase de fabricação. Entretanto a quantidade de possibilidades de soluções a ser avaliadas cresce de maneira exponencial. Avaliar o custo de uma dessas soluções através da síntese em hardware é uma tarefa extremamente custosa. E mesmo a alternativa de síntese em alto nível não permite sintetizar todas as soluções e atender ao time-to-market. O uso de técnicas de aprendizado de máquina na DSE reduz a quantidade de sínteses e simulações necessárias ao estimar novos valores a partir de um conjunto de dados de treinamento, possibilitando alcançar elevadas taxas de acurácia e atender ao time-to-market. Nesse trabalho é apresentada uma investigação do estado da arte do uso da técnica de aprendizado de máquina para a DSE de arquiteturas computacionais para sistemas embarcados.


Keywords


exploração de espaço de projeto, sistemas embarcados, aprendizado de máquina

References


AL-RAKHAMI, M. et al. An ensemble learning approach for accurate energy load prediction in residential buildings. IEEE Access, IEEE, v. 7, p. 48328–48338, 2019.

ARM. Arm big.little technologies. . Accesso em: 14/04/2020.

AUSTIN, T.; LARSON, E.; ERNST, D. Simplescalar: An infrastructure for computer system modeling. Computer, IEEE, n. 2, p. 59–67, 2002.

BELLARD, F. Qemu, a fast and portable dynamic translator. In: USENIX Annual Technical Conference, FREENIX Track. [S.l.: s.n.], 2005. v. 41, p. 46.

BINKERT, N. et al. The gem5 simulator. ACM SIGARCH Computer Architecture News, ACM, v. 39, n. 2, p. 1–7, 2011.

BREUGHE, M. B.; EYERMAN, S.; EECKHOUT, L. Mechanistic analytical modeling of superscalar in-order processor performance. ACM Transactions on Architecture and Code Optimization (TACO), ACM, v. 11, n. 4, p. 50, 2015.

BUTKO, A. et al. Accuracy evaluation of gem5 simulator system. In: IEEE. 7th International Workshop on Reconfigurable and Communication-Centric Systems-on-Chip (ReCoSoC). [S.l.], 2012. p. 1–7.

CHIŞ, R. et al. Multi-objective optimization for an enhanced multi-core sniper simulator. 2018.

COMPTON, K.; HAUCK, S. Reconfigurable computing: a survey of systems and software. ACM Computing Surveys (csuR), ACM, v. 34, n. 2, p. 171–210, 2002.

CONG, J. et al. Accelerator-rich architectures: Opportunities and progresses. In: ACM. Proceedings of the 51st Annual Design Automation Conference. [S.l.], 2014. p. 1–6.

DAMAK, B. et al. Design space exploration for customized asymmetric heterogeneous mpsoc. In: IEEE. 2014 17th Euromicro Conference on Digital System Design. [S.l.], 2014. p. 50–57.

DIDONA, D. et al. Enhancing performance prediction robustness by combining analytical modeling and machine learning. In: ACM. Proceedings of the 6th ACM/SPEC International conference on performance engineering. [S.l.], 2015. p. 145–156.

DOPPA, J. R.; ROSCA, J.; BOGDAN, P. Autonomous design space exploration of computing systems for sustainability: Opportunities and challenges. IEEE Design & Test, IEEE, v. 36, n. 5, p. 35–43, 2019

DUTTA, B.; ADHINARAYANAN, V.; FENG, W.-c. Gpu power prediction via ensemble machine learning for dvfs space exploration. In: ACM. Proceedings of the 15th ACM International Conference on Computing Frontiers. [S.l.], 2018. p. 240–243.

FERRETTI, L.; ANSALONI, G.; POZZI, L. Cluster-based heuristic for high level synthesis design space exploration. IEEE Transactions on Emerging Topics in Computing, IEEE, 2018.

FULKERSON, B. Machine learning, neural and statistical classification. [S.l.]: Taylor & Francis, 1995.

GREATHOUSE, J. L.; LOH, G. H. Machine learning for performance and power modeling of heterogeneous systems. In: IEEE. 2018 IEEE/ACM International Conference on Computer-Aided Design (ICCAD). [S.l.], 2018. p. 1–6.

GUI, C.-Y. et al. A survey on graph processing accelerators: Challenges and opportunities. Journal of Computer Science and Technology, Springer, v. 34, n. 2, p. 339–371, 2019.

GUO, Q. et al. Effective and efficient microprocessor design space exploration using unlabeled design configurations. In: Twenty-Second International Joint Conference on Artificial Intelligence. [S.l.: s.n.], 2011.

HARTENSTEIN, R. Coarse grain reconfigurable architecture (embedded tutorial). In: Proceedings of the 2001 Asia and South Pacific Design Automation Conference. New York,NY, USA: ACM, 2001. (ASP-DAC ’01), p. 564–570. ISBN 0-7803-6634-4.

HILL, M. D.; MARTY, M. R. Amdahl’s law in the multicore era. Computer, IEEE, v. 41, n. 7, p. 33–38, 2008.

ÏPEK, E. et al. Efficiently exploring architectural design spaces via predictive modeling. [S.l.]: ACM, 2006. v. 41.

KANG, S.; KUMAR, R. Magellan: a search and machine learning-based framework for fast multi-core design space exploration and optimization. In: IEEE. 2008 Design, Automation and Test in Europe. [S.l.], 2008. p. 1432–1437.

KHAN, S. et al. Using predictive modeling for cross-program design space exploration in multicore systems. In: IEEE COMPUTER SOCIETY. Proceedings of the 16th International Conference on Parallel Architecture and Compilation Techniques. [S.l.], 2007. p. 327–338.

KIM, R. G.; DOPPA, J. R.; PANDE, P. P. Machine learning for design space exploration and optimization of manycore systems. In: IEEE. 2018 IEEE/ACM International Conference on Computer-Aided Design (ICCAD). [S.l.], 2018. p. 1–6.

LEUPERS, R. et al. Software compilation techniques for heterogeneous embedded multi-core systems. In: Handbook of Signal Processing Systems. [S.l.]: Springer, 2019. p.1021–1062

LI, D.; YAO, S.; WANG, Y. Processor design space exploration via statistical sampling and semi-supervised ensemble learning. IEEE Access, IEEE, v. 6, p. 25495–25505, 2018.

LIN, C. et al. An efficient and effective performance estimation method for dse. In: IEEE. 2016 International Symposium on VLSI Design, Automation and Test (VLSI-DAT). [S.l.], 2016. p. 1–4.

LIU, F. et al. A survey on edge computing systems and tools. Proceedings of the IEEE,IEEE, v. 107, n. 8, p. 1537–1562, 2019.

MADSEN, J. et al. Multi-objective design space exploration of embedded system platforms. In: SPRINGER. IFIP Working Conference on Distributed and Parallel Embedded Systems. [S.l.], 2006. p. 185–194.

MAGNUSSON, P. S. et al. Simics: A full system simulation platform. Computer, IEEE, v. 35, n. 2, p. 50–58, 2002.

MALAZGIRT, G. A.; YURDAKUL, A. Prenaut: Design space exploration for embedded symmetric multiprocessing with various on-chip architectures. Journal of Systems Architecture, Elsevier, v. 72, p. 3–18, 2017.

MARIANI, G. et al. Oscar: An optimization methodology exploiting spatial correlation in multicore design spaces. IEEE Transactions on Computer-Aided Design of Integrated Circuits and Systems, IEEE, v. 31, n. 5, p. 740–753, 2012.

MENG, P. et al. Adaptive threshold non-pareto elimination: Re-thinking machine learning for system level design space exploration on fpgas. In: EDA CONSORTIUM. Proceedings

of the 2016 Conference on Design, Automation & Test in Europe. [S.l.], 2016. p. 918–923.

MUNG, P. S.; PHYU, S. Effective analytics on healthcare big data using ensemble learning. In: IEEE.2020 IEEE Conference on Computer Applications (ICCA). [S.l.], 2020.p. 1–4.

NIKOLOV, N.; NAKOV, O. Creating architecture and software of embedded systems with constrained resources and their communication to the iot cloud. In: IEEE. 2019 X National Conference with International Participation (ELECTRONICA). [S.l.], 2019.p. 1–4

NVIDIA. Tegra mobile processors. 2019. . Accesso em: 29-06-2019.

O’NEAL, K.; BRISK, P. Predictive modeling for cpu, gpu, and fpga performance and power consumption: A survey. In: IEEE. 2018 IEEE Computer Society Annual Symposium on VLSI (ISVLSI). [S.l.], 2018. p. 763–768.

OVP. Open Vitual Platform. 2008. . Acesso em: 2019-12-09.

OZISIKYILMAZ, B.; MEMIK, G.; CHOUDHARY, A. Efficient system design space exploration using machine learning techniques. In: ACM. Proceedings of the 45th annual design automation conference. [S.l.], 2008. p. 966–969.

PISCITELLI, R.; PIMENTEL, A. D. Design space pruning through hybrid analysis in system-level design space exploration. In: EDA CONSORTIUM. Proceedings of the Conference on Design, Automation and Test in Europe. [S.l.], 2012. p. 781–786.

REDDY, D. et al. Bridging functional heterogeneity in multicore architectures. ACM SIGOPS Operating Systems Review, ACM New York, NY, USA, v. 45, n. 1, p. 21–33, 2011.

SAMSUNG. The Samsung Reference Platform. 2019. . Accesso em: 02/06/2019.

SARMA, S.; DUTT, N. Cross-layer exploration of heterogeneous multicore processor configurations. In: IEEE. 2015 28th International Conference on VLSI Design. [S.l.], 2015. p. 147–152.

SCHAFER, B. C.; WANG, Z. High-level synthesis design space exploration: Past, present and future. IEEE Transactions on Computer-Aided Design of Integrated Circuits and Systems, IEEE, 2019

SCOGLAND, T. et al. Asymmetric interactions in symmetric multi-core systems: analysis, enhancements and evaluation. In: IEEE. SC’08: Proceedings of the 2008 ACM/IEEE Conference on Supercomputing. [S.l.], 2008. p. 1–12.

SHAHID, A. et al. Ac-dse: approximate computing for the design space exploration of reconfigurable mpsocs. Journal of Circuits, Systems and Computers, World Scientific, v. 27, n. 09, p. 1850145, 2018.

SINGH, G. et al. Napel: Near-memory computing application performance prediction via ensemble learning. In: ACM. Proceedings of the 56th Annual Design Automation Conference 2019. [S.l.], 2019. p. 27.

SOTIRIOU-XANTHOPOULOS, E. et al. An integrated exploration and virtual platform framework for many-accelerator heterogeneous systems. ACM Transactions on Embedded Computing Systems (TECS), ACM, v. 15, n. 3, p. 43, 2016.

STOCKMAN, M.; AWAD, M. Predicting microarchitectural power using interval based hierarchical support vector machine. In: IEEE. 2010 International Conference on Energy Aware Computing. [S.l.], 2010. p. 1–3.

WANG, C. et al. Reconfigurable hardware accelerators: Opportunities, trends, and challenges. arXiv preprint arXiv:1712.04771, 2017.

WANG, L.; SKADRON, K. Lumos+: Rapid, pre-rtl design space exploration on accelerator-rich heterogeneous architectures with reconfigurable logic. In: IEEE. 2016 IEEE 34th International Conference on Computer Design (ICCD). [S.l.], 2016. p. 328–335.

YOURST, M. T. Ptlsim: A cycle accurate full system x86-64 microarchitectural simulator. In: IEEE. 2007 IEEE International Symposium on Performance Analysis of Systems & Software. [S.l.], 2007. p. 23–34.




DOI: https://doi.org/10.34117/bjdv6n5-215

Refbacks

  • There are currently no refbacks.