ABSTRACT
Our goals are to address challenges such as latency, scalability, throughput and heterogeneous data sources of streaming analytics and deep learning pipelines in science sensors and medical imaging applications. We present a prototype Intelligent Parallel Distributed Streaming Framework (IPDSF) that is capable of distributed streaming processing as well as performing distributed deep training in batch mode. IPDSF is designed to run streaming Artificial Intelligent (AI) analytic tasks using data parallelism including partitions of multiple streams of short time sensing data and high-resolution 3D medical images, and fine grain tasks distribution. We will show the implementation of IPDSF for two real world applications, (i) an Air Quality Index based on near real time streaming of aerosol Lidar backscatter and (ii) data generation of Covid-19 Computing Tomography (CT) scans using deep learning. We evaluate the latency, throughput, scalability, and quantitative evaluation of training and prediction compared against a baseline single instance. As the results, IPDSF scales to process thousands of streaming science sensors in parallel for Air Quality Index application. IPDSF uses novel 3D conditional Generative Adversarial Network (cGAN) training using parallel distributed Graphic Processing Units (GPU) nodes to generate realistic 3D high resolution Computed Tomography scans of Covid-19 patient lungs. We will show that IPDSF can reduce cGAN training time linearly with the number of GPUs.
- A. A. Awan, A. Jain, C.-H. Chu, H. Subramoni, and D. Panda, “Communication Profiling and Characterization of Deep Learning Workloads on Clusters with High-Performance Interconnects,” in Hot Interconnects 26 (HotI ’19), August 2019.Google ScholarCross Ref
- “AmazonKinesis,”https://aws.amazon.com/kinesis/data-streams/.Google Scholar
- Anthony, Q., Awan, A. A., Jain, A., Subramoni, H., & Panda, D. K. D. K. (2020). Efficient training of semantic image segmentation on summit using horovod and MVAPICH2-GDR. 2020 IEEE International Parallel and Distributed Processing Symposium Workshops (IPDPSW), 1015–1023. IEEE..Google ScholarCross Ref
- B. Yadranjiaghdam, S. Yasrobi, and N. Tabrizi, “Developing a real-time data analytics framework for twitter streaming data,” in 2017 IEEE International Congress on Big Data (BigData Congress), June 2017, pp. 329–336.Google ScholarCross Ref
- D. Tiwari, S. Gupta, and S. S. Vazhkudai, “Lazy checkpointing: Exploiting temporal locality in failures to mitigate checkpointing overheads on extreme-scale systems,” International Conference on Dependable Systems and Networks (DSN), 2014.Google ScholarDigital Library
- H. Ye, Z. Wu, R.-W. Zhao, X. Wang, Y.-G. Jiang, and X. Xue, “Evaluating two-stream cnn for video classification,” in Proceedings of the 5th ACM on International Conference on Multimedia Retrieval, ser. ICMR ’15. New York, NY, USA: ACM, 2015, pp. 435–442. [Online]. Available:http://doi.acm.org/10.1145/2671188.2749406Google ScholarDigital Library
- Ichinose, A., Takefusa, A., Nakada, H., & Oguchi, M. (2017). A study of a video analysis framework using Kafka and spark streaming. 2017 IEEE International Conference on Big Data (Big Data), 2396–2401. IEEE.Google ScholarCross Ref
- J. Read, F. Perez-Cruz, and A. Bifet, “Deep learning in partially labeled data streams,” in Proceedings of the 30th Annual ACM Symposium on Applied Computing, ser. SAC ’15. New York, NY, USA: ACM, 2015, pp. 954–959. [Online].Available:http://doi.acm.org/10.1145/2695664.2695871Google ScholarDigital Library
- Tanaka, K., Arikawa, Y., Ito, T., Morita, K., Nemoto, N., Miura, F., … Sakamoto, T. (2020). Communication-efficient distributed deep learning with GPU-FPGA heterogeneous computing. 2020 IEEE Symposium on High-Performance Interconnects (HOTI), 43–46. IEEEGoogle ScholarCross Ref
- Tudoran, R., Nano, O., Santos, I., Costan, A., Soncu, H., Bougé, L., & Antoniu, G. (2014). JetStream: Enabling high performance event streaming across cloud data-centers. Proceedings of the 8th ACM International Conference on Distributed Event-Based Systems - DEBS ’14. New York, New York, USA: ACM Press.Google ScholarDigital Library
- Marcu, O.-C., Costan, A., Antoniu, G., Perez-Hernandez, M., Nicolae, B., Tudoran, R., & Bortoli, S. (2018). KerA: Scalable Data Ingestion for Stream Processing. 2018 IEEE 38th International Conference on Distributed Computing Systems (ICDCS), 1480–1485. IEEE.Google Scholar
- A. Sergeev and M. Del Balso, Horovod: Fast and Easy Distributed Deep Learning in TensorFlow, arXiv:1802.05799v3, Feb. 21, 2018.Google Scholar
- Tiwari, D., Gupta, S., Gallarno, G., Rogers, J., & Maxwell, D. (2015). Reliability lessons learned from GPU experience with the Titan supercomputer at Oak Ridge leadership computing facility. Proceedings of the International Conference for High Performance Computing, Networking, Storage and Analysis on - SC ’15. New York, New York, USA: ACM Press.Google ScholarDigital Library
- J. Sleeman, Z. Yang, V.Caicedo, M. Halem, B. Demoz, R. Delgado, “A Deep Machine Learning Approach for LIDAR Based Boundary Layer Height Detection”, International Geoscience and Remote Sensing Symposium (IGARSS).Google Scholar
- P.Nguyen and M.Halem “Deep Learning Models for Predicting CO2 Flux Employing Multivariate Time Series” SIGKDD MileTS, Alaska 2019.Google Scholar
- J. Mangalagiri, D. Chapman, A. Gangopadhyay, Y. Yesha, J. Galita, S. Menon, Y. Yesha, B. Saboury, M. Morris, P. Nguyen, “Toward Generating Synthetic CT Volumes using a 3D-Conditional Generative Adversarial Network”, The 2020 International Conference on Computational Science and Computational Intelligence Symposium on Health Informatics and Medical Systems (CSCI-ISHI), Las Vegas 2020.Google ScholarCross Ref
- Nokleby, M., Raja, H., & Bajwa, W. U. (2020). Scaling-up distributed processing of data streams for machine learning. Proceedings of the IEEE. Institute of Electrical and Electronics Engineers, 1–29.Google ScholarCross Ref
- Phuong Nguyen, Samit Shivadekar, Sai Sree Chukkapalli, Milton Halem. Satellite Data Fusion of Multiple Observed XCO2 using Compressive Sensing and Deep Learning. IEEE IGARSS 2020Google Scholar
- Y. Mirsky, T. Mahler, I. Shelef, and Y. Elovici, “CT-GAN: Malicious tampering of 3D medical imagery using deep learning,” In 28th {USENIX} Security Symposium ({USENIX} Security 19) (pp. 461-478), 2019.Google Scholar
- Efremov, M. A., Kholod, I. I., & Kolpaschikov, M. A. (2021). Java Federated Learning framework architecture. 2021 IEEE Conference of Russian Young Researchers in Electrical and Electronic Engineering (ElConRus), 306–309.Google ScholarCross Ref
- Kim, D., Park, H., & Choi, J. K. (2021). Optimal load allocation for coded distributed computation in heterogeneous clusters. IEEE Transactions on Communications, 69(1), 44–58.Google ScholarCross Ref
- Reyes-Anastacio, H. G., Gonzalez-Compean, J. L., Sosa-Sosa, V. J., Carretero, J., & Garcia-Blas, J. (2020). Kulla, a container-centric construction model for building infrastructure-agnostic distributed and parallel applications. The Journal of Systems and Software, 168(110665), 110665.Google ScholarCross Ref
Index Terms
- An Intelligent Parallel Distributed Streaming Framework for near Real-time Science Sensors and High-Resolution Medical Images
Recommendations
Analysis of high-resolution reconstruction of medical images based on deep convolutional neural networks in lung cancer diagnostics
Highlights- The diagnostic effect of 64-slice spiral CT and MRI high-resolution images based on deep CNN in lung cancer is studied.
Abstract Background and objectiveTo study the diagnostic effect of 64-slice spiral CT and MRI high-resolution images based on deep convolutional neural networks(CNN) in lung cancer.
MethodsIn this ...
Towards Scalable Parallel Training of Deep Neural Networks
MLHPC'17: Proceedings of the Machine Learning on HPC EnvironmentsWe propose a new framework for parallelizing deep neural network training that maximize the amount of data that is ingested by the training algorithm. Our proposed framework called Livermore Tournament Fast Batch Learning (LTFB) targets large-scale data ...
High Resolution Medical Image Segmentation Using Data-Swapping Method
Medical Image Computing and Computer Assisted Intervention – MICCAI 2019AbstractDeep neural network models used for medical image segmentation are large because they are trained with high-resolution three-dimensional (3D) images. Graphics processing units (GPUs) are widely used to accelerate training. However, the memory on a ...
Comments