Avaliação de Cluster Raspberry Pi para Execução de Aplicações de Análise de Imagens Microscópicas Médicas
Resumo
A área de saúde e, em especial, a patologia clinica com análise automatizada de imagens microscópicas é um campo com demanda crescente por poder computacional. Uma barreira para a utilização eficaz de computação nesse cenário é a indisponibilidade de recursos computacionais suficientes, devido aos altos custos relacionados. Dessa forma, nesse artigo, avaliamos a utilização de uma arquitetura de baixo custo e gasto energético formada por equipamentos Raspberry Pi 2 para construção de aglomerados de computadores com 64 cores e 16GB RAM para utilização em aplicações médicas em análise de imagens microscópicas. Nossa avaliação tem como objetivo identificar o potencial custo benefício dessa plataforma em comparação com outros processadores, levando em consideração tempo de execução, custo do hardware e gasto energético. Os resultados experimentais mostraram que a utilização de um cluster de Raspberrys é 10× e 2× mais rápido que a execução da aplicação, respectivamente, em máquinas Core2Duo e I7. Mesmo em sua capacidade máxima, o cluster foi energeticamente mais econômico do que os demais processadores. Dessa forma, o Raspberry Pi 2 mostrou-se uma excelente e promissora plataforma para execução de nossa classe de aplicações alvo.
Referências
The Message Passing Interface (MPI).
Beynon, M. D., Kurc, T., Catalyurek, U., Chang, C., Sussman, A., and Saltz, J. (2001). Distributed processing of very large datasets with DataCutter. Parallel Comput., 27(11):1457–1478.
Bradski, G. (2000). The OpenCV Library. Dr. Dobb’s Journal of Software Tools.
Butler, H. (2012). Intel core i7-3770k power consumption. http://www.bit-tech.net/hardware/2012/04/23/intel-core-i7-3770k-review/8.
Chang, C., Moon, B., Acharya, A., Shock, C., Sussman, A., and Saltz, J. H. (1997). Titan: A High-Performance Remote Sensing Database. In Proceedings of the Thirteenth International Conference on Data Engineering, ICDE ’97, pages 375–384, Washington, DC, USA. IEEE Computer Society.
Cooper, L. A. D., Kong, J., Gutman, D. A., Wang, F., Cholleti, S. R., Pan, T. C., Widener, P. M., Sharma, A., Mikkelsen, T., Flanders, A. E., Rubin, D. L., Meir, E. G. V., Kurc, T. M., Moreno, C. S., Brat, D. J., and Saltz, J. H. (2010). An integrative approach for in silico glioma research. IEEE Trans Biomed Eng., 57(10):2617–2621.
Corporation, S. (2016a). Cartao de memoria sandisk extreme microsdhc/microsdxc uhsi. https://www.sandisk.com.br/home/memory-cards/microsd-cards/extreme-microsd.
Corporation, S. (2016b). Cartao de memoria sandisk ultra microsdhc/microsdxc uhs-i para cameras. https://www.sandisk.com.br/home/memory-cards/microsd-cards/ultra-microsd-for-cameras.
Cox, S. J., Cox, J. T., Boardman, R. P., Johnston, S. J., Scott, M., and O’Brien, N. S. (2013). Iridis-pi: a low-cost, compact demonstration cluster. Cluster Computing, 17(2):349–358.
DELL (2016). Ecommerce dell. http://www.dell.com.br.
FOUNDATION., R. P. (2015). Raspberry pi 2 model b. https://www.raspberrypi.org/products/raspberry-pi-2-model-b/.
Klie, H., Bangerth,W., Gai, X., Wheeler, M. F., Stoffa, P. L., Sen, M. K., Parashar, M., C¸ atalyurek, U¨ . V., Saltz, J. H., and Kurc¸, T. M. (2006). Models, methods and middleware for grid-enabled multiphysics oil reservoir management. Eng. Comput. (Lond.), 22(3-4):349–370.
Kong, J., Cooper, L. A. D., Wang, F., Gao, J., Teodoro, G., Mikkelsen, T., Schniederjan, M. J., Moreno, C. S., Saltz, J. H., and Brat, D. J. (2013). Machine-based morphologic analysis of glioblastoma using whole-slide pathology images uncovers clinically relevant molecular correlates. PLoS ONE.
Kurc, T., Qi, X., Wang, D., Wang, F., Teodoro, G., Cooper, L., Nalisnik, M., Yang, L., Saltz, J., and Foran, D. J. (2015). Scalable analysis of big pathology image data cohorts using efficient methods and high-performance computing strategies. BMC Bioinformatics, 16(1):1.
Kurc¸, T. M., C¸ atalyu¨rek, U¨ . V., Zhang, X., Saltz, J. H., Martino, R., Wheeler, M. F., Peszynska, M., Sussman, A., Hansen, C., Sen, M. K., Seifoullaev, R., Stoffa, P. L., Torres-Verd´ın, C., and Parashar, M. (2005). A simulation and data analysis system for large-scale, data-driven oil reservoir simulation studies. Concurrency - Practice and Experience, 17(11):1441–1467.
Page, M. (2009). Cpu power consumption tests. http://www.pcstats.com/articleview.cfm?articleid=2395&page=2.
Parashar, M., Matossian, V., Bangerth, W., Klie, H., Rutt, B., Kurc¸, T. M., C¸ atalyurek, U¨ . V., Saltz, J. H., and Wheeler, M. F. (2005). Towards Dynamic Data-Driven Optimization of Oil Well Placement. In International Conference on Computational Science (2), pages 656–663.
Raspian (2016). Raspian. http://www.raspian.org/.
Shock, C. T., Chang, C., Moon, B., Acharya, A., Davis, L., Saltz, J., and Sussman, A. (1998). The design and evaluation of a high-performance earth science database. Parallel Computing, 24(1):65 – 89.
Teodoro, G., Kurc, T., Andrade, G., Kong, J., Ferreira, R., and Saltz, J. (2015). Application performance analysis and efficient execution on systems with multi-core CPUs, GPUs and MICs: a case study with microscopy image analysis. International Journal of High Performance Computing Applications.
Teodoro, G., Kurc¸, T. M., Kong, J., Cooper, L. A. D., and Saltz, J. H. (2014). Comparative Performance Analysis of Intel (R) Xeon Phi (TM), GPU, and CPU: A Case Study from Microscopy Image Analysis. In 2014 IEEE 28th International Parallel and Distributed Processing Symposium, May 19-23, 2014, pages 1063–1072.
Teodoro, G., Pan, T., Kurc, T. M., Cooper, L., Kong, J., and Saltz, J. H. (2012). A fast parallel implementation of queue-based morphological reconstruction using GPUs. Emory University, Center for Comprehensive Informatics, CCI-TR-2012-2.
Teodoro, G., Pan, T., Kurc, T. M., Kong, J., Cooper, L. A., Podhorszki, N., Klasky, S., and Saltz, J. H. (2013a). High-throughput Analysis of Large Microscopy Image Datasets on CPU-GPU Cluster Platforms. In IPDPS ’13: Proceedings of the 2013 IEEE International Symposium on Parallel and Distributed Processing.
Teodoro, G., Pan, T., Kurc, T. M., Kong, J., Cooper, L. A., and Saltz, J. H. (2013b). Efficient Irregular Wavefront Propagation Algorithms on Hybrid CPU-GPU machines. Parallel Computing, 39(4-5):189–211.
Vincent, L. (1993). Morphological grayscale reconstruction in image analysis: Applications and efficient algorithms. IEEE Transactions on Image Processing, 2:176–201.
Beynon, M. D., Kurc, T., Catalyurek, U., Chang, C., Sussman, A., and Saltz, J. (2001). Distributed processing of very large datasets with DataCutter. Parallel Comput., 27(11):1457–1478.
Bradski, G. (2000). The OpenCV Library. Dr. Dobb’s Journal of Software Tools.
Butler, H. (2012). Intel core i7-3770k power consumption. http://www.bit-tech.net/hardware/2012/04/23/intel-core-i7-3770k-review/8.
Chang, C., Moon, B., Acharya, A., Shock, C., Sussman, A., and Saltz, J. H. (1997). Titan: A High-Performance Remote Sensing Database. In Proceedings of the Thirteenth International Conference on Data Engineering, ICDE ’97, pages 375–384, Washington, DC, USA. IEEE Computer Society.
Cooper, L. A. D., Kong, J., Gutman, D. A., Wang, F., Cholleti, S. R., Pan, T. C., Widener, P. M., Sharma, A., Mikkelsen, T., Flanders, A. E., Rubin, D. L., Meir, E. G. V., Kurc, T. M., Moreno, C. S., Brat, D. J., and Saltz, J. H. (2010). An integrative approach for in silico glioma research. IEEE Trans Biomed Eng., 57(10):2617–2621.
Corporation, S. (2016a). Cartao de memoria sandisk extreme microsdhc/microsdxc uhsi. https://www.sandisk.com.br/home/memory-cards/microsd-cards/extreme-microsd.
Corporation, S. (2016b). Cartao de memoria sandisk ultra microsdhc/microsdxc uhs-i para cameras. https://www.sandisk.com.br/home/memory-cards/microsd-cards/ultra-microsd-for-cameras.
Cox, S. J., Cox, J. T., Boardman, R. P., Johnston, S. J., Scott, M., and O’Brien, N. S. (2013). Iridis-pi: a low-cost, compact demonstration cluster. Cluster Computing, 17(2):349–358.
DELL (2016). Ecommerce dell. http://www.dell.com.br.
FOUNDATION., R. P. (2015). Raspberry pi 2 model b. https://www.raspberrypi.org/products/raspberry-pi-2-model-b/.
Klie, H., Bangerth,W., Gai, X., Wheeler, M. F., Stoffa, P. L., Sen, M. K., Parashar, M., C¸ atalyurek, U¨ . V., Saltz, J. H., and Kurc¸, T. M. (2006). Models, methods and middleware for grid-enabled multiphysics oil reservoir management. Eng. Comput. (Lond.), 22(3-4):349–370.
Kong, J., Cooper, L. A. D., Wang, F., Gao, J., Teodoro, G., Mikkelsen, T., Schniederjan, M. J., Moreno, C. S., Saltz, J. H., and Brat, D. J. (2013). Machine-based morphologic analysis of glioblastoma using whole-slide pathology images uncovers clinically relevant molecular correlates. PLoS ONE.
Kurc, T., Qi, X., Wang, D., Wang, F., Teodoro, G., Cooper, L., Nalisnik, M., Yang, L., Saltz, J., and Foran, D. J. (2015). Scalable analysis of big pathology image data cohorts using efficient methods and high-performance computing strategies. BMC Bioinformatics, 16(1):1.
Kurc¸, T. M., C¸ atalyu¨rek, U¨ . V., Zhang, X., Saltz, J. H., Martino, R., Wheeler, M. F., Peszynska, M., Sussman, A., Hansen, C., Sen, M. K., Seifoullaev, R., Stoffa, P. L., Torres-Verd´ın, C., and Parashar, M. (2005). A simulation and data analysis system for large-scale, data-driven oil reservoir simulation studies. Concurrency - Practice and Experience, 17(11):1441–1467.
Page, M. (2009). Cpu power consumption tests. http://www.pcstats.com/articleview.cfm?articleid=2395&page=2.
Parashar, M., Matossian, V., Bangerth, W., Klie, H., Rutt, B., Kurc¸, T. M., C¸ atalyurek, U¨ . V., Saltz, J. H., and Wheeler, M. F. (2005). Towards Dynamic Data-Driven Optimization of Oil Well Placement. In International Conference on Computational Science (2), pages 656–663.
Raspian (2016). Raspian. http://www.raspian.org/.
Shock, C. T., Chang, C., Moon, B., Acharya, A., Davis, L., Saltz, J., and Sussman, A. (1998). The design and evaluation of a high-performance earth science database. Parallel Computing, 24(1):65 – 89.
Teodoro, G., Kurc, T., Andrade, G., Kong, J., Ferreira, R., and Saltz, J. (2015). Application performance analysis and efficient execution on systems with multi-core CPUs, GPUs and MICs: a case study with microscopy image analysis. International Journal of High Performance Computing Applications.
Teodoro, G., Kurc¸, T. M., Kong, J., Cooper, L. A. D., and Saltz, J. H. (2014). Comparative Performance Analysis of Intel (R) Xeon Phi (TM), GPU, and CPU: A Case Study from Microscopy Image Analysis. In 2014 IEEE 28th International Parallel and Distributed Processing Symposium, May 19-23, 2014, pages 1063–1072.
Teodoro, G., Pan, T., Kurc, T. M., Cooper, L., Kong, J., and Saltz, J. H. (2012). A fast parallel implementation of queue-based morphological reconstruction using GPUs. Emory University, Center for Comprehensive Informatics, CCI-TR-2012-2.
Teodoro, G., Pan, T., Kurc, T. M., Kong, J., Cooper, L. A., Podhorszki, N., Klasky, S., and Saltz, J. H. (2013a). High-throughput Analysis of Large Microscopy Image Datasets on CPU-GPU Cluster Platforms. In IPDPS ’13: Proceedings of the 2013 IEEE International Symposium on Parallel and Distributed Processing.
Teodoro, G., Pan, T., Kurc, T. M., Kong, J., Cooper, L. A., and Saltz, J. H. (2013b). Efficient Irregular Wavefront Propagation Algorithms on Hybrid CPU-GPU machines. Parallel Computing, 39(4-5):189–211.
Vincent, L. (1993). Morphological grayscale reconstruction in image analysis: Applications and efficient algorithms. IEEE Transactions on Image Processing, 2:176–201.
Publicado
04/07/2016
Como Citar
RAMOS, Rafael M.; RALHA, Célia; TEODORO, George.
Avaliação de Cluster Raspberry Pi para Execução de Aplicações de Análise de Imagens Microscópicas Médicas. In: SEMINÁRIO INTEGRADO DE SOFTWARE E HARDWARE (SEMISH), 43. , 2016, Porto Alegre.
Anais [...].
Porto Alegre: Sociedade Brasileira de Computação,
2016
.
p. 1795-1806.
ISSN 2595-6205.
DOI: https://doi.org/10.5753/semish.2016.9528.
