Patrick O'Leary
Research Subjects
Research Interests
Patrick O'Leary has a total of 32 co-authors. Here are the top co-authors by number of shared publications:
- Brad Whitlock
Intelligent Light
3 shared publications - Burlen Loring
Lawrence Berkeley National Laboratory
2 shared publications - David Pugmire
Computer Science and Mathematics Division, Oak Ridge National Laboratory, Oak Ridge, TN, USA
2 shared publications - David Thompson
Kitware, Inc
2 shared publications - E. Wes Bethel
Lawrence Berkeley National Laboratory
2 shared publications - Earl Duque
Intelligent Light
2 shared publications
Latest Publications
- 5 May 2022
- Published by Springer Nature
- p. 439-460
Proximity Portability and in Transit, M-to-N Data Partitioning and Movement in SENSEI
- 5 May 2022
- Published by Springer Nature
- p. 439-460
Proximity Portability and in Transit, M-to-N Data Partitioning and Movement in SENSEI
- 5 May 2022
- Published by Springer Nature
- p. 281-306
The SENSEI Generic In Situ Interface: Tool and Processing Portability at Scale
- 5 May 2022
- Published by Springer Nature
- p. 281-306
The SENSEI Generic In Situ Interface: Tool and Processing Portability at Scale
- Open Access
Sandtank-ML: An Educational Tool at the Interface of Hydrology and Machine Learning
Hydrologists and water managers increasingly face challenges associated with extreme climatic events. At the same time, historic datasets for modeling contemporary and future hydrologic conditions are increasingly inadequate. Machine learning is one promising technological tool for navigating the challenges of understanding and managing contemporary hydrological systems. However, in addition to the technical challenges associated with effectively leveraging ML for understanding subsurface hydrological processes, practitioner skepticism and hesitancy surrounding ML presents a significant barrier to adoption of ML technologies among practitioners. In this paper, we discuss an educational application we have developed—Sandtank-ML—to be used as a training and educational tool aimed at building user confidence and supporting adoption of ML technologies among water managers. We argue that supporting the adoption of ML methods and technologies for subsurface hydrological investigations and management requires not only the development of robust technologic tools and approaches, but educational strategies and tools capable of building confidence among diverse users.
Open AccessSandtank-ML: An Educational Tool at the Interface of Hydrology and Machine Learning
Hydrologists and water managers increasingly face challenges associated with extreme climatic events. At the same time, historic datasets for modeling contemporary and future hydrologic conditions are increasingly inadequate. Machine learning is one promising technological tool for navigating the challenges of understanding and managing contemporary hydrological systems. However, in addition to the technical challenges associated with effectively leveraging ML for understanding subsurface hydrological processes, practitioner skepticism and hesitancy surrounding ML presents a significant barrier to adoption of ML technologies among practitioners. In this paper, we discuss an educational application we have developed—Sandtank-ML—to be used as a training and educational tool aimed at building user confidence and supporting adoption of ML technologies among water managers. We argue that supporting the adoption of ML methods and technologies for subsurface hydrological investigations and management requires not only the development of robust technologic tools and approaches, but educational strategies and tools capable of building confidence among diverse users.
- Open Access
Building Near-Real-Time Processing Pipelines with the Spark-MPI Platform
Advances in detectors and computational technologies provide new opportunities for applied research and the fundamental sciences. Concurrently, dramatic increases in the three Vs (Volume, Velocity, and Variety) of experimental data and the scale of computational tasks produced the demand for new real-time processing systems at experimental facilities. Recently, this demand was addressed by the Spark-MPI approach connecting the Spark data-intensive platform with the MPI high-performance framework. In contrast with existing data management and analytics systems, Spark introduced a new middleware based on resilient distributed datasets (RDDs), which decoupled various data sources from high-level processing algorithms. The RDD middleware significantly advanced the scope of data-intensive applications, spreading from SQL queries to machine learning to graph processing. Spark-MPI further extended the Spark ecosystem with the MPI applications using the Process Management Interface. The paper explores this integrated platform within the context of online ptychographic and tomographic reconstruction pipelines.
Open AccessBuilding Near-Real-Time Processing Pipelines with the Spark-MPI Platform
Advances in detectors and computational technologies provide new opportunities for applied research and the fundamental sciences. Concurrently, dramatic increases in the three Vs (Volume, Velocity, and Variety) of experimental data and the scale of computational tasks produced the demand for new real-time processing systems at experimental facilities. Recently, this demand was addressed by the Spark-MPI approach connecting the Spark data-intensive platform with the MPI high-performance framework. In contrast with existing data management and analytics systems, Spark introduced a new middleware based on resilient distributed datasets (RDDs), which decoupled various data sources from high-level processing algorithms. The RDD middleware significantly advanced the scope of data-intensive applications, spreading from SQL queries to machine learning to graph processing. Spark-MPI further extended the Spark ecosystem with the MPI applications using the Process Management Interface. The paper explores this integrated platform within the context of online ptychographic and tomographic reconstruction pipelines.