Search

This study presents a novel approach to using a gated recurrent unit (GRU) model, a deep neural network, to predict turbulent flows in a Lagrangian framework. The emerging velocity field is predicted based on experimental data from a strained turbulent flow, which was initially a nearly homogeneous isotropic turbulent flow at the measurement area. The distorted turbulent flow has a Taylor microscale Reynolds number in the range of 100 < R e λ < 152 before creating the strain and is strained with a mean strain rate of 4 s − 1 in the Y direction. The measurement is conducted in the presence of gravity consequent to the actual condition, an effect that is usually neglected and has not been investigated in most numerical studies. A Lagrangian particle tracking technique is used to extract the flow characterizations. It is used to assess the capability of the GRU model to forecast the unknown turbulent flow pattern affected by distortion and gravity using spatiotemporal input data. Using the flow track's location (spatial) and time (temporal) highlights the model's superiority. The suggested approach provides the possibility to predict the emerging pattern of the strained turbulent flow properties observed in many natural and artificial phenomena. In order to optimize the consumed computing, hyperparameter optimization (HPO) is used to improve the GRU model performance by 14–20%. Model training and inference run on the high-performance computing (HPC) JUWELS-BOOSTER and DEEP-DAM systems at the Jülich Supercomputing Centre, and the code speed-up on these machines is measured. The proposed model produces accurate predictions for turbulent flows in the Lagrangian view with a mean absolute error (MAE) of 0.001 and an R 2 score of 0.993. [ABSTRACT FROM AUTHOR]

Continuously increasing data volumes from multiple sources, such as simulation and experimental measurements, demand efficient algorithms for an analysis within a realistic timeframe. Deep learning models have proven to be capable of understanding and analyzing large quantities of data with high accuracy. However, training them on massive datasets remains a challenge and requires distributed learning exploiting High-Performance Computing systems. This study presents a comprehensive analysis and comparison of three well-established distributed deep learning frameworks—Horovod, DeepSpeed, and Distributed Data Parallel by PyTorch—with a focus on their runtime performance and scalability. Additionally, the performance of two data loaders, the native PyTorch data loader and the DALI data loader by NVIDIA, is investigated. To evaluate these frameworks and data loaders, three standard ResNet architectures with 50, 101, and 152 layers are tested using the ImageNet dataset. The impact of different learning rate schedulers on validation accuracy is also assessed. The novel contribution lies in the detailed analysis and comparison of these frameworks and data loaders on the state-of-the-art Jülich Wizard for European Leadership Science (JUWELS) Booster system at the Jülich Supercomputing Centre, using up to 1024 A100 NVIDIA GPUs in parallel. Findings show that the DALI data loader significantly reduces the overall runtime of ResNet50 from more than 12 h on 4 GPUs to less than 200 s on 1024 GPUs. The outcomes of this work highlight the potential impact of distributed deep learning using efficient tools on accelerating scientific discoveries and data-driven applications. [ABSTRACT FROM AUTHOR]

The COVID-19 pandemic shed light on the need for quick diagnosis tools in healthcare, leading to the development of several algorithmic models for disease detection. Though these models are relatively easy to build, their training requires a lot of data, storage, and resources, which may not be available for use by medical institutions or could be beyond the skillset of the people who most need these tools. This paper describes a data analysis and machine learning platform that takes advantage of high-performance computing infrastructure for medical diagnosis support applications. This platform is validated by re-training a previously published deep learning model (COVID-Net) on new data, where it is shown that the performance of the model is improved through large-scale hyperparameter optimisation that uncovered optimal training parameter combinations. The per-class accuracy of the model, especially for COVID-19 and pneumonia, is higher when using the tuned hyperparameters (healthy: 96.5%; pneumonia: 61.5%; COVID-19: 78.9%) as opposed to parameters chosen through traditional methods (healthy: 93.6%; pneumonia: 46.1%; COVID-19: 76.3%). Furthermore, training speed-up analysis shows a major decrease in training time as resources increase, from 207 min using 1 node to 54 min when distributed over 32 nodes, but highlights the presence of a cut-off point where the communication overhead begins to affect performance. The developed platform is intended to provide the medical field with a technical environment for developing novel portable artificial-intelligence-based tools for diagnosis support. [ABSTRACT FROM AUTHOR]