Robert Dietrich
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Research Subjects
Research Interests
Robert Dietrich has a total of 9 co-authors. Here are the top co-authors by number of shared publications:
- Guido Juckeland
Helmholtz-Zentrum Dresden-Rossendorf, Dresden, Saxony
3 shared publications Ronny Tschüter
Technische Universität Dresden
3 shared publicationsAndreas Knüpfer
Technische Universität Dresden
2 shared publicationsFelix Schmitt
Center for Information Services and High Performance Computing (ZIH), Technische Universität Dresden, Germany
1 shared publication- Felix Schmitt
Technische Universität Dresden
1 shared publication Frank Winkler
Technische Universität Dresden
1 shared publication
Latest Publications
- 23 May 2021
- Published by Springer Nature
- p. 169-182
Enabling Performance Analysis of Kokkos Applications with Score-P
- 23 May 2021
- Published by Springer Nature
- p. 169-182
Enabling Performance Analysis of Kokkos Applications with Score-P
- 20 October 2017
- Published by Springer Nature
- p. 457-476
Analyzing Offloading Inefficiencies in Scalable Heterogeneous Applications
- 20 October 2017
- Published by Springer Nature
- p. 457-476
Analyzing Offloading Inefficiencies in Scalable Heterogeneous Applications
- 1 August 2016
- Published by SAGE Publications in The International Journal of High Performance Computing Applications
- Vol. 31 (6) , 485-498
Scalable critical-path analysis and optimization guidance for hybrid MPI-CUDA applications
The use of accelerators in heterogeneous systems is an established approach in designing petascale applications. Today, Compute Unified Device Architecture (CUDA) offers a rich programming interface for GPU accelerators but requires developers to incorporate several layers of parallelism on both the CPU and the GPU. From this increasing program complexity emerges the need for sophisticated performance tools. This work contributes by analyzing hybrid MPI-CUDA programs for properties based on wait states, such as the critical path, a metric proven to identify application bottlenecks effectively. We developed a tool to construct a dependency graph based on an execution trace and the inherent dependencies of the programming models CUDA and Message Passing Interface (MPI). Thereafter, it detects wait states and attributes blame to responsible activities. Together with the property of being on the critical path, we can identify activities that are most viable for optimization. To evaluate the global impact of optimizations to critical activities, we predict the program execution using a graph-based performance projection. The developed approach has been demonstrated with suitable examples to be both scalable and correct. Furthermore, we establish a new categorization of CUDA inefficiency patterns ensuing from the dependencies between CUDA activities.
- 1 August 2016
- Published by SAGE Publications in The International Journal of High Performance Computing Applications
- Vol. 31 (6) , 485-498
Scalable critical-path analysis and optimization guidance for hybrid MPI-CUDA applications
The use of accelerators in heterogeneous systems is an established approach in designing petascale applications. Today, Compute Unified Device Architecture (CUDA) offers a rich programming interface for GPU accelerators but requires developers to incorporate several layers of parallelism on both the CPU and the GPU. From this increasing program complexity emerges the need for sophisticated performance tools. This work contributes by analyzing hybrid MPI-CUDA programs for properties based on wait states, such as the critical path, a metric proven to identify application bottlenecks effectively. We developed a tool to construct a dependency graph based on an execution trace and the inherent dependencies of the programming models CUDA and Message Passing Interface (MPI). Thereafter, it detects wait states and attributes blame to responsible activities. Together with the property of being on the critical path, we can identify activities that are most viable for optimization. To evaluate the global impact of optimizations to critical activities, we predict the program execution using a graph-based performance projection. The developed approach has been demonstrated with suitable examples to be both scalable and correct. Furthermore, we establish a new categorization of CUDA inefficiency patterns ensuing from the dependencies between CUDA activities.
- 28 July 2016
- Published by Springer Nature
- p. 67-83
Evaluation of Tool Interface Standards for Performance Analysis of OpenACC and OpenMP Programs
- 28 July 2016
- Published by Springer Nature
- p. 67-83
Evaluation of Tool Interface Standards for Performance Analysis of OpenACC and OpenMP Programs
- 1 January 2015
- Published by Springer Nature
- p. 161-173
Integrating Critical-Blame Analysis for Heterogeneous Applications into the Score-P Workflow
- 1 January 2015
- Published by Springer Nature
- p. 161-173
Integrating Critical-Blame Analysis for Heterogeneous Applications into the Score-P Workflow