Print this Page


Our GCOE members work on many different software projects, almost all of them are open source and will be listed here soon.


Benchmark Suite for Heterogeneous Implementations of FFTs | MPI-CBG
This is a simple and easy extensible benchmark system to answer the question, which FFT library performs best under which conditions. Conditions are given by compute architecture, inplace or outplace as well as real or complex transforms, data precision, and so on.
Issue TrackerLicense: Apache-2.0Publication: gearshifft - The FFT Benchmark Suite for Heterogeneous Platforms


A particle-in-cell code for GPGPUs | HZDR
PIConGPU is a fully relativistic, many GPGPU, 3D3V particle-in-cell (PIC) code. The Particle-in-Cell algorithm is a central tool in plasma physics. It describes the dynamics of a plasma by computing the motion of electrons and ions in the plasma based on Maxwell's equations.
Issue Tracker


Abstraction Library for Parallel Kernel Acceleration | HZDR
The alpaka library is a header-only C++11 abstraction library for accelerator development. Its aim is to provide performance portability across accelerators through the abstraction (not hiding!) of the underlying levels of parallelism.
Issue Tracker


C++ User interface for the Platform independent Library Alpaka | HZDR
cupla is a simple user interface for the platform independent parallel kernel acceleration library alpaka. It follows a similar concept as the NVIDIA® CUDA® API by providing a software layer to manage accelerator devices. alpaka is used as backend for cupla.
Issue Tracker


In Situ Animation of Accelerated Computations | HZDR
Many computations like physics or biologists simulations these days run on accelerated hardware like CUDA GPUs or Intel Xeon Phi, which are itself distributed in a big compute cluster communicating over MPI. The goal of ISAAC is to visualize this data without the need to download it to the host while using the high computation speed of the accelerator.
Issue Tracker


Memory Allocator for Many Core Architectures | HZDR
This project provides a framework for fast memory managers on many core accelerators. Currently, it supports NVIDIA GPUs of compute capability sm_20 or higher through the ScatterAlloc algorithm.
Issue Tracker


Graph Approach for Highly Generic Communication Schemes Based on Adaptive Topologies | HZDR
Graybat is a C++ library that presents a graph-based communication approach, which enables a mapping of algorithms to communication patterns and further a mapping of these communication patterns to varying hardware architectures. Therefore, a flexible and configurable communication approach for parallel and distributed applications. These mappings are established as an intermediate layer between an application and communication libraries and are dynamically adptable during run-time.
Issue Tracker


Specification of HPC hardware and program components | HZDR
An extensible framework to describe HPC hardware as well as large-scale scientific simulations. Dodo uses a hardware model, a worker model, data elements, and algorithms which are all described using graph-based data structures. These data structures are annotated with custom meta data to describe each layer in detail. Domain decompositions are expressed through graph mappings. These mappings combine the different models to form a comprehensive view of a simulation run.
Issue Tracker


PARAllel Tracer for Arbitrary X-ray Interaction and Scattering | HZDR
This is a simulation framework for X-Ray Tracing with libPMacc.
Issue Tracker


library for Library-independent FFTs | HZDR
This library contains a generic FFT interface that relies on C++11 features and template metaprogramming to do lots of compile time checks on the validity of the input. It also dynamically compiles only the used library code, which makes it possible to include libraries headers that are not installed on the system, as those are not used in that case.
Issue Tracker


Shrink-Wrap Phase Reconstruction Algorithm | HZDR
CUDA accelerated Phase Reconstruction Algorithm
Issue Tracker


Reconstruction of activity from PET data using rays | HZDR
raptr stands for reconstruction of activity from PET data using rays. It is a code developed at the HZDR for iterative algebraic image reconstruction of activity densities from PET measurements. raptr uses CUDA devices for acceleration and runs parallely across multiple nodes.
Issue Tracker


High performance Amplified Spontaneous Emission on GPU | HZDR
CUDA accelerated simulation of simulating the gain in laser media used in high-power laser systems.
Issue Tracker


Simulation of strongly coupled plasma using GPUs | HZDR
SCPonGPU is a code to simulate laser cooling of ions in storage rings and particle traps. In order to simulate the strong coupling of ions in these scenarios, the Coulomb interaction of all particles with each other is considered. This approach is called a molecular dynamics simulation, short MD, and has a computational complexity of NxN (with N being the number of particles). Since this product can be quite large, we simulate the particle-particle interaction on GPUs using CUDA, where we can exploit a highly parallel architecture to accelerate the simulation.
Issue Tracker


Distributed Interactive Visualization and Exploration of large datasets | HZDR
Use pyDive to work with homogeneous, n-dimensional arrays that are too big to fit into your local machine's memory. pyDive provides containers whose elements are distributed across a cluster or stored in a large hdf5/adios-file if the cluster is still too small. All computation and data-access is then done in parallel (even on GPUs!) by the cluster nodes in the background. If you feel like working with numpy arrays pyDive has reached the goal!
Issue Tracker


Fork of CUDA GPU memtest | HZDR
Fork and updated (maintained) version of the famous cuda_memtest tool from the University of Illinois/NCSA. Supports recent Nvidia hardware and fixes dozens of bugs.
Issue Tracker


Open Standard for Particle-Mesh Data Files | HZDR, LBNL, LLNL, MPI-CBG, TU Dresden, IOQ Jena
The openPMD standard, short for open standard for particle-mesh data files is not a file format per se. It is a standard for meta data and naming schemes. Its used throughout several HPC projects including PIConGPU to exchange self-describing data between high-bandwidth simulations, experiments, post- and in situ processing frameworks.
Issue Tracker


Simple Parallel file output Library for Accumulating Simulation data using Hdf5 | ZIH, HZDR
libSplash aims at developing a HDF5-based I/O library for HPC simulations. It is created as an easy-to-use frontend for the standard HDF5 library with support for MPI processes in a cluster environment. While the standard HDF5 library provides detailed low-level control, libSplash simplifies tasks commonly found in large-scale HPC simulations, such as iterative computations and MPI distributed processes.
Issue Tracker

Permanent link to this article: