All Outputs (38)

SYCL compute kernels for ExaHyPE (2024)
Presentation / Conference Contribution
Loi, C. M., Bockhorst, H., & Weinzierl, T. (2024, March). SYCL compute kernels for ExaHyPE. Presented at 2024 SIAM Conference on Parallel Processing for Scientific Computing (PP), Baltimore, MD

We discuss three SYCL realisations of a simple Finite Volume scheme over multiple Cartesian patches. The realisation flavours differ in the way how they map the compute steps onto loops and tasks: We compare an implementation that is exclusively usin... Read More about SYCL compute kernels for ExaHyPE.

Detrimental task execution patterns in mainstream OpenMP runtimes (2024)
Presentation / Conference Contribution
Weinzierl, T., Tuft, A., & Klemm, M. (2024, September). Detrimental task execution patterns in mainstream OpenMP runtimes. Presented at IWOMP 2024, Perth, Australia

The OpenMP API offers both task-based and data-parallel concepts to scientific computing. While it provides descriptive and prescriptive annotations, it is in many places deliberately unspecific how to implement its annotations. As the predomina... Read More about Detrimental task execution patterns in mainstream OpenMP runtimes.

Compiler support for semi-manual AoS-to-SoA conversions with data views (2024)
Presentation / Conference Contribution
Radtke, P., & Weinzierl, T. (2024, September). Compiler support for semi-manual AoS-to-SoA conversions with data views. Presented at PPAM 2024 - 15th International Conference on Parallel Processing & Applied Mathematics, Ostrava, Czech Republic

The C programming language and its cousins such as C++ stipulate the static storage of sets of structured data: Developers have to commit to one, invariant data model -- typically a structure-of-arrays (SoA) or an array-of-structs (AoS) -- unles... Read More about Compiler support for semi-manual AoS-to-SoA conversions with data views.

A multiscale optimisation algorithm for shape and material reconstruction from a single X-ray image (2024)
Presentation / Conference Contribution
Westmacott, H., Ivrissimtzis, I., & Weinzierl, T. (2024, January). A multiscale optimisation algorithm for shape and material reconstruction from a single X-ray image. Presented at ICIGP 2024: The 7th International Conference on Image and Graphics Processing, Beijing, China

We produce thickness and bone to soft tissue ratio estimations from a single, 2D medical X-ray image. For this, we simulate the scattering of the rays through a model of the object and embed this simulation into an optimiser which iteratively adjusts... Read More about A multiscale optimisation algorithm for shape and material reconstruction from a single X-ray image.

Efficient GPU Offloading with OpenMP for a Hyperbolic Finite Volume Solver on Dynamically Adaptive Meshes (2023)
Presentation / Conference Contribution
Wille, M., Weinzierl, T., Brito Gadeschi, G., & Bader, M. (2023, December). Efficient GPU Offloading with OpenMP for a Hyperbolic Finite Volume Solver on Dynamically Adaptive Meshes. Presented at ISC High Performance 2023, Hamburg

We identify and show how to overcome an OpenMP bottleneck in the administration of GPU memory. It arises for a wave equation solver on dynamically adaptive block-structured Cartesian meshes, which keeps all CPU threads busy and allows all of them to... Read More about Efficient GPU Offloading with OpenMP for a Hyperbolic Finite Volume Solver on Dynamically Adaptive Meshes.

Doubt and Redundancy Kill Soft Errors---Towards Detection and Correction of Silent Data Corruption in Task-based Numerical Software (2021)
Presentation / Conference Contribution
Samfass, P., Weinzierl, T., Reinarz, A., & Bader, M. (2021, November). Doubt and Redundancy Kill Soft Errors---Towards Detection and Correction of Silent Data Corruption in Task-based Numerical Software. Presented at Supercomputing 21 - FTXS Workshop - 2021 IEEE/ACM 11th Workshop on Fault Tolerance for HPC at eXtreme Scale (FTXS), St Louis, MO

Resilient algorithms in high-performance computing are subject to rigorous non-functional constraints. Resiliency must not increase the runtime, memory footprint or I/O demands too significantly. We propose a task-based soft error detection scheme th... Read More about Doubt and Redundancy Kill Soft Errors---Towards Detection and Correction of Silent Data Corruption in Task-based Numerical Software.

Task inefficiency patterns for a wave equation solver (2021)
Presentation / Conference Contribution
Schulz, H., Brito Gadeschi, G., Rudyy, O., & Weinzierl, T. (2021, December). Task inefficiency patterns for a wave equation solver. Presented at IWOMP 2021, Bristol

teaMPI---replication-based resiliency without the (performance) pain (2020)
Presentation / Conference Contribution
Samfass, P., Weinzierl, T., Hazelwood, B., & Bader, M. (2020, December). teaMPI---replication-based resiliency without the (performance) pain. Presented at ISC High Performance, Frankfurt

In an era where we can not afford to checkpoint frequently, replication is a generic way forward to construct numerical simulations that can continue to run even if hardware parts fail. Yet, replication often is not employed on larger scales, as naïv... Read More about teaMPI---replication-based resiliency without the (performance) pain.

Lazy Stencil Integration in multigrid algorithms (2019)
Presentation / Conference Contribution
Murray, C., & Weinzierl, T. (2019, December). Lazy Stencil Integration in multigrid algorithms. Presented at 13th INTERNATIONAL CONFERENCE ON PARALLEL PROCESSING AND APPLIED MATHEMATICS, Bialystok, Poland

Simulation of tsunamis with the exascale hyperbolic PDE engine ExaHyPE (2018)
Presentation / Conference Contribution
Rannabauer, L., Haas, S., Charrier, D., Weinzierl, T., & Bader, M. (2018, December). Simulation of tsunamis with the exascale hyperbolic PDE engine ExaHyPE. Presented at Environmental Informatics: Techniques and Trends. Adjunct Proceedings of the 32nd edition of the EnviroInfo

An experience report on (auto-)tuning of mesh-based PDE solvers on shared memory systems (2018)
Presentation / Conference Contribution
Charrier, D. E., & Weinzierl, T. (2018, March). An experience report on (auto-)tuning of mesh-based PDE solvers on shared memory systems. Presented at PPAM 2017, Lublin, Poland

With the advent of manycore systems, shared memory parallelisation has gained importance in high performance computing. Once a code is decomposed into tasks or parallel regions, it becomes crucial to identify reasonable grain sizes, i.e. minimum prob... Read More about An experience report on (auto-)tuning of mesh-based PDE solvers on shared memory systems.

Fast DEM collision checks on multicore nodes (2018)
Presentation / Conference Contribution
Krestenitis, K., Weinzierl, T., & Koziara, T. (2017, September). Fast DEM collision checks on multicore nodes. Presented at 12th International Conference on Parallel Processing and Applied Mathematics (PPAM) 2017, Lublin, Poland

Many particle simulations today rely on spherical or analytical particle shape descriptions. They find non-spherical, triangulated particle models computationally infeasible due to expensive collision detections. We propose a hybrid collision detecti... Read More about Fast DEM collision checks on multicore nodes.

A Case Study for a New Invasive Extension of Intel's Threading Building Blocks (2018)
Presentation / Conference Contribution
Schreiber, M., & Weinzierl, T. (2018, January). A Case Study for a New Invasive Extension of Intel's Threading Building Blocks. Presented at HiPEAC 2018 - 3rd COSH Workshop on Co-Scheduling of HPC Applications, Manchester, England

We study codes deploying multiple MPI ranks to one node where each rank is parallelised with TBB. A static assignment of cores to ranks here is disadvantageous if the load is not perfectly balanced, the runtime is subject to fluctuations or one MPI r... Read More about A Case Study for a New Invasive Extension of Intel's Threading Building Blocks.

Form Follows Function - Do algorithms and applications challenge or drag behind the hardware evolution? (2016)
Presentation / Conference Contribution
Weinzierl, T. (2016, June). Form Follows Function - Do algorithms and applications challenge or drag behind the hardware evolution?. Presented at ISC High Performance 2016, Frankfurt am Main, Germany

Exascale roadmaps are dominated by predictions on hardware trends. At the same time, hardware-software co-design is a frequently cited phrase. It suggests that software development has an impact on the hardware evolution. Is this assumption valid? If... Read More about Form Follows Function - Do algorithms and applications challenge or drag behind the hardware evolution?.

On-the-fly memory compression for multibody algorithms (2016)
Presentation / Conference Contribution
Eckhardt, W., Glas, R., Korzh, D., Wallner, S., & Weinzierl, T. (2015, September). On-the-fly memory compression for multibody algorithms. Presented at International Conference on Parallel Computing (ParCo) 2015, Edinburgh, Scotland

Memory and bandwidth demands challenge developers of particle-based codes that have to scale on new architectures, as the growth of concurrency outperforms improvements in memory access facilities, as the memory per core tends to stagnate, and as com... Read More about On-the-fly memory compression for multibody algorithms.

A Contact Detection Code using Triangles for Non-Spherical Particle Simulations (2016)
Presentation / Conference Contribution
Krestenitis, K., Weinzierl, T., & Koziara, T. (2023, March). A Contact Detection Code using Triangles for Non-Spherical Particle Simulations. Presented at 24th Conference on Computational Mechanics (ACME-UK 2016), Cardiff, UK

We present a novel DEM (discrete element method) code with explicit time stepping. DEM codes simulate billions of small particles that interact with each other primarily through collisions. Different to state-of-the-art codes, we rely on triangulated... Read More about A Contact Detection Code using Triangles for Non-Spherical Particle Simulations.

Integration of FULLSWOF2D and PeanoClaw: Adaptivity and Local Time-stepping for Complex Overland Flows (2015)
Presentation / Conference Contribution
Unterweger, K., Wittmann, R., Neumann, P., Weinzierl, T., & Bungartz, H.-J. (2014, October). Integration of FULLSWOF2D and PeanoClaw: Adaptivity and Local Time-stepping for Complex Overland Flows. Presented at 3rd International Workshop on Computational Engineering CE 2014., Stuttgart, Germany

We propose to couple our adaptive mesh refinement software PeanoClaw with existing solvers for complex overland flows that are tailored to regular Cartesian meshes. This allows us to augment them with spatial adaptivity and local time-stepping withou... Read More about Integration of FULLSWOF2D and PeanoClaw: Adaptivity and Local Time-stepping for Complex Overland Flows.

Adaptivity - and Regular Cartesian Patches for the Shallow Water Equations and Vectorisation of an Augmented Riemann Solver for the Shallow Water Equations (2014)
Presentation / Conference Contribution
Bader, M., & Weinzierl, T. (2014, December). Adaptivity - and Regular Cartesian Patches for the Shallow Water Equations and Vectorisation of an Augmented Riemann Solver for the Shallow Water Equations. Paper presented at SIAM Annual Meeting, Chicago, IL

Hardware-aware block size tailoring on adaptive spacetree grids for shallow water waves (2014)
Presentation / Conference Contribution
Weinzierl, T., Wittmann, R., Unterweger, K., Bader, M., Breuer, A., & Rettenberger, S. (2014, January). Hardware-aware block size tailoring on adaptive spacetree grids for shallow water waves. Presented at HiStencils 2014 - 1st International Workshop on High-Performance Stencil Computations, Vienna, Austria

Spacetrees are a popular formalism to describe dynamically adaptive Cartesian grids. Though they directly yield an adaptive spatial discretisation, i.e. a mesh, it is often more efficient to augment them by regular Cartesian blocks embedded into the... Read More about Hardware-aware block size tailoring on adaptive spacetree grids for shallow water waves.

Advanced Computing (2013)
Presentation / Conference Contribution
(2013, December). Advanced Computing. Presented at Institut für Informatik, Technische Universität München