David Boehme
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Previous Organizations / Institutes
The organizations / institutes from the scholar's other articles in the past ten years
Research Subjects
Research Interests
David Boehme has a total of 30 co-authors. Here are the top co-authors by number of shared publications:
- Olga Pearce
Lawrence Livermore National Laboratory, Livermore, USA
6 shared publications - Anthony Skjellum
Tennessee Technological University
3 shared publications - Bengisu Elis
Technical University of Munich, Garching, Germany
2 shared publications - Derek Schafer
University of New Mexico
2 shared publications - Jason Burmark
Lawrence Livermore National Laboratory, Livermore, USA
2 shared publications - Martin Schulz
Technische Universität München
2 shared publications
Latest Publications
- Open Access
- 25 January 2026
- Published by Association for Computing Machinery (ACM)
- p. 215-227
GPU Partitioning, Power, and Performance of the AMD MI300A
- 25 January 2026
- Published by Association for Computing Machinery (ACM)
- p. 215-227
Open AccessGPU Partitioning, Power, and Performance of the AMD MI300A
- Open Access
Leveraging Caliper and Benchpark to Analyze MPI Communication Patterns: Insights from AMG2023, Kripke, and Laghos
We introduce ``communication regions'' into the widely used Caliper HPC profiling tool. A communication region is an annotation enabling capture of metrics about the data being communicated (including statistics of these metrics), and metrics about the MPI processes involved in the communications, something not previously possible in Caliper. We explore the utility of communication regions with three representative modeling and simulation applications, AMG2023, Kripke, and Laghos, all part of the comprehensive Benchpark suite that includes Caliper annotations. Enhanced Caliper reveals detailed communication behaviors. Using Caliper and Thicket in tandem, we create new visualizations of MPI communication patterns, including halo exchanges. Our findings reveal communication bottlenecks and detailed behaviors, indicating significant utility of the special-regions addition to Caliper. The comparative scaling behavior of both CPU and GPU oriented systems are shown; we are able to look at different regions within a given application, and see how scalability and message-traffic metrics differ.
Open AccessLeveraging Caliper and Benchpark to Analyze MPI Communication Patterns: Insights from AMG2023, Kripke, and Laghos
We introduce ``communication regions'' into the widely used Caliper HPC profiling tool. A communication region is an annotation enabling capture of metrics about the data being communicated (including statistics of these metrics), and metrics about the MPI processes involved in the communications, something not previously possible in Caliper. We explore the utility of communication regions with three representative modeling and simulation applications, AMG2023, Kripke, and Laghos, all part of the comprehensive Benchpark suite that includes Caliper annotations. Enhanced Caliper reveals detailed communication behaviors. Using Caliper and Thicket in tandem, we create new visualizations of MPI communication patterns, including halo exchanges. Our findings reveal communication bottlenecks and detailed behaviors, indicating significant utility of the special-regions addition to Caliper. The comparative scaling behavior of both CPU and GPU oriented systems are shown; we are able to look at different regions within a given application, and see how scalability and message-traffic metrics differ.
- 26 August 2024
- Published by Springer Nature
- p. 107-120
A Mechanism to Generate Interception Based Tools for HPC Libraries
- 26 August 2024
- Published by Springer Nature
- p. 107-120
A Mechanism to Generate Interception Based Tools for HPC Libraries
- Open Access
MPI Implementation Profiling for Better Application Performance
While application profiling has been a mainstay in the HPC community for years, profiling of MPI and other communication middleware has not received the same degree of exploration. This paper adds to the discussion of MPI profiling, contributing two general-purpose profiling methods as well as practical applications of these methods to an existing implementation. The ability to detect performance defects in MPI codes using these methods increases the potential of further research and development in communication optimization.
Open AccessMPI Implementation Profiling for Better Application Performance
While application profiling has been a mainstay in the HPC community for years, profiling of MPI and other communication middleware has not received the same degree of exploration. This paper adds to the discussion of MPI profiling, contributing two general-purpose profiling methods as well as practical applications of these methods to an existing implementation. The ability to detect performance defects in MPI codes using these methods increases the potential of further research and development in communication optimization.
- Open Access
- 18 January 2024
- Published by Association for Computing Machinery (ACM)
- p. 1-11
Non-Blocking GPU-CPU Notifications to Enable More GPU-CPU Parallelism
- 18 January 2024
- Published by Association for Computing Machinery (ACM)
- p. 1-11
Open AccessNon-Blocking GPU-CPU Notifications to Enable More GPU-CPU Parallelism
- Open Access
- 7 August 2023
- Published by Association for Computing Machinery (ACM)
- p. 281-293
Thicket: Seeing the Performance Experiment Forest for the Individual Run Trees
Thicket is an open-source Python toolkit for Exploratory Data Analysis (EDA) of multi-run performance experiments. It enables an understanding of optimal performance configuration for large-scale application codes. Most performance tools focus on a single execution (e.g., single platform, single measurement tool, single scale). Thicket bridges the gap to convenient analysis in multi-dimensional, multi-scale, multi-architecture, and multi-tool performance datasets by providing an interface for interacting with the performance data. Thicket has a modular structure composed of three components. The first component is a data structure for multi-dimensional performance data, which is composed automatically on the portable basis of call trees, and accommodates any subset of dimensions present in the dataset. The second is the metadata, enabling distinction and sub-selection of dimensions in performance data. The third is a dimensionality reduction mechanism, enabling analysis such as computing aggregated statistics on a given data dimension. Extensible mechanisms are available for applying analyses (e.g., top-down on Intel CPUs), data science techniques (e.g., K-means clustering from scikit-learn), modeling performance (e.g., Extra-P), and interactive visualization. We demonstrate the power and flexibility of Thicket through two case studies, first with the open-source RAJA Performance Suite on CPU and GPU clusters and another with a large physics simulation run on both a traditional HPC cluster and an AWS Parallel Cluster instance.
- 7 August 2023
- Published by Association for Computing Machinery (ACM)
- p. 281-293
Open AccessThicket: Seeing the Performance Experiment Forest for the Individual Run Trees
Thicket is an open-source Python toolkit for Exploratory Data Analysis (EDA) of multi-run performance experiments. It enables an understanding of optimal performance configuration for large-scale application codes. Most performance tools focus on a single execution (e.g., single platform, single measurement tool, single scale). Thicket bridges the gap to convenient analysis in multi-dimensional, multi-scale, multi-architecture, and multi-tool performance datasets by providing an interface for interacting with the performance data. Thicket has a modular structure composed of three components. The first component is a data structure for multi-dimensional performance data, which is composed automatically on the portable basis of call trees, and accommodates any subset of dimensions present in the dataset. The second is the metadata, enabling distinction and sub-selection of dimensions in performance data. The third is a dimensionality reduction mechanism, enabling analysis such as computing aggregated statistics on a given data dimension. Extensible mechanisms are available for applying analyses (e.g., top-down on Intel CPUs), data science techniques (e.g., K-means clustering from scikit-learn), modeling performance (e.g., Extra-P), and interactive visualization. We demonstrate the power and flexibility of Thicket through two case studies, first with the open-source RAJA Performance Suite on CPU and GPU clusters and another with a large physics simulation run on both a traditional HPC cluster and an AWS Parallel Cluster instance.
- 1 May 2012
- Published by Institute of Electrical and Electronics Engineers (IEEE)
- p. 2538-2541
Characterizing Load and Communication Imbalance in Large-Scale Parallel Applications
Load or communication imbalance prevents many codes from taking advantage of the parallelism available on modern supercomputers. We present two scalable methods to highlight imbalance in parallel programs: The first method identifies delays that inflict wait states at subsequent synchronization points, and attributes their costs in terms of resource waste to the original cause. The second method combines knowledge of the critical path with traditional parallel profiles to derive a set of compact performance indicators that help answer a variety of important performance-analysis questions, such as identifying load imbalance, quantifying the impact of imbalance on runtime, and characterizing resource consumption. Both methods employ a highly scalable parallel replay of event traces, making them a suitable analysis instrument for massively parallel MPI programs with tens of thousands of processes.
- 1 May 2012
- Published by Institute of Electrical and Electronics Engineers (IEEE)
- p. 2538-2541
Characterizing Load and Communication Imbalance in Large-Scale Parallel Applications
Load or communication imbalance prevents many codes from taking advantage of the parallelism available on modern supercomputers. We present two scalable methods to highlight imbalance in parallel programs: The first method identifies delays that inflict wait states at subsequent synchronization points, and attributes their costs in terms of resource waste to the original cause. The second method combines knowledge of the critical path with traditional parallel profiles to derive a set of compact performance indicators that help answer a variety of important performance-analysis questions, such as identifying load imbalance, quantifying the impact of imbalance on runtime, and characterizing resource consumption. Both methods employ a highly scalable parallel replay of event traces, making them a suitable analysis instrument for massively parallel MPI programs with tens of thousands of processes.