Publications of Information Services and Computing (from 2015)

Involved institute: Information Services and Computing (from 2015)
Mit ausgewählten Publikationen vor FWC
"Online First" included
Without submitted and only approved publications
Only approved publications

119 Publications
Research Data Management to increase research quality
Konrad, U.
The publication is an invited plenary lecture at the national workshop on documentation and information in Jakarta, Indonesia.
Research data play a fundamental role in todays open science driven research. In addition the software used to generate, process or analyse the data has to be considered for the whole research and publication process to ensure, that results are findable, accessible, interoperable, re-usable and re-producible. The lecture describes the challenges, tasks, solutions and organization structures to meet the challenges for the library and data center at the Helmholtz-Center Dresden-Rossendorf and beyond.
Keywords: Open Science, Research Data Management, Software Management, Data Center, Library
  • Invited lecture (Conferences)
    National Workshop on Documentation and Information Data, Information and Knowledge, 25.-26.10.2017, Jakarta, Indonesia
    DOI: 10.5281/zenodo.1040289

Downloads:

Registration No. 26206 - Permalink


Wissenschaftliche Software – Anspruch und Realität im Forschungsprozess
Konrad, U.ORC
Wissenschaftliche Software ist heute unverzichtbares Werkzeug im Forschungsprozess, sie ist Voraussetzung für die Nachvollziehbarkeit der (publizierten) Ergebnisse und in vielen Fällen auch selbst ein Ergebnis, das publiziert, genutzt und langfristig bewahrt werden muss. Publikationen bestehen künftig häufig aus zitierfähigen Texten, Daten und Software und müssen entsprechend konsistent behandelt werden, dies ist eine Herausforderung auch für die Bibliotheken. Daraus ergeben sich eine Reihe von Fragestellungen und Aufgaben für die gute wissenschaftliche Praxis im Umfeld der „Offenen Wissenschaft“ (Open Science). Diese Fragen werden in dem Vortrag diskutiert.
Ausgangspunkt ist die Frage, welche Kategorien wissenschaftlicher Software es gibt und was für Rollen diese im Forschungsprozess spielen. Mit der voranschreitenden Digitalisierung von Forschung und Lehre steigt die Abhängigkeit von Software-Lösungen. Die grundlegenden Prinzipien der „guten wissenschaftliche Praxis“ wie Nachvollziehbarkeit, Reproduzierbarkeit, Transparenz und Qualitätssicherung müssen auch bei der Entwicklung und Nutzung von wissenschaftlicher Software angewandt werden. Allerdings gibt es eine Reihe von Besonderheiten für den Umgang mit Software, da diese im Vergleich zu Veröffentlichungen und Daten einem meist kontinuierlichen Entwicklungsprozess unterliegt und im Kontext spezifischer Entwicklungs- und Laufzeitumgebungen zu betrachten sind.
Darüber hinaus sind die Publikation, Nachnutzbarkeit und Verwertung von Software zentrale Herausforderungen. Die Zitation von Quellcode, die Open Source Software-Entwicklung, die Bereitstellung forschungsnaher Infrastrukturen für Entwicklung und Test sowie die Lizenzen und rechtliche Aspekte der Softwarenachnutzung sind noch nicht umfassend in der wissenschaftlichen Praxis realisiert. Darüber hinaus fehlen auch Anreizsysteme für eine nachhaltige Softwareentwicklung in der Forschung.
Für viele dieser Fragen gibt es Lösungsansätze und „best practice“ Beispiele, auf die eingegangen wird. Seit einiger Zeit gibt es dazu internationale und nationale Initiativen wie u.a. die Software Carpentry (1998, US), das Software Sustainability Institute (2008, GB) und sciforge (2014, D). Des Weiteren unterstützt die Deutsche Forschungsgemeinschaft (DFG) die Entwicklung dieses Gebietes u.a. mit dem Programm ”Research Software Sustainability”.
Keywords: Wissenschaftliche Software, Softwareentwicklung, Publikation, Repositorien, Infrastruktur, Bibliothek, Scientific software, software development, publication, repository, infrastructure
  • Lecture (Conference)
    Internationale Open Access Tage 2017, 11.-13.09.2017, Dresden, Deutschland
    DOI: 10.5281/zenodo.1040289

Registration No. 26025 - Permalink


Detecting Threatening States in Laser Beams
Kelling, J.; Juckeland, G.
This poster presents our approach to automatic detection of critical failure states in the pulsed Petawatt laser systems DRACO and PENELOPE, used for investigations of exotic states of matter and medical applications. The beam shape is controlled to avoid high destructive energy densities. However, randomly occurring states threatening the device must be detected between pulses and trigger an interlock in the device firing at 10Hz.

The states we are aiming to detect are rare; thus, training data for this category is scarce. To address this, we present two approaches: First, to identify regions of interest based on physical properties of the system and apply a convolutional neural network (CNN) to to identify true positives. Secondly, using CNN-based image segmentation to localize and classify regions of interest.
Keywords: image classification, deep learning, smart laser operation
  • Poster
    Deep Learning Bootcamp 2017, 21.-25.08.2017, Dresden, Deutschland

Registration No. 25962 - Permalink


Efficient Parallel Monte-Carlo Simulations for Large-Scale Studies of Surface Growth Processes
Kelling, J.; Ódor, G.; Weigel, M.; Gemming, G.
Lattice Monte Carlo methods are used to investigate far from and out-of-equilibrium systems, including surface growth, spin systems and solid mixtures. Such studies require observations of large systems over long times scales, to allow structures to grow over orders of magnitude, which necessitates massively parallel simulations. This talk presents work done to address the problem of parallel processing introducing correlations in Monte Carlo updates. Studies of the effect of correlations on scaling and dynamical properties of surface growth systems and related lattice gases is investigated further by comparing results obtained by correlation-free and intrinsically correlated simulations. Where the latter, based on a stochastic cellular automaton approach, are of interest because of their high efficiency. The primary subject of study is the Kardar–Parisi–Zhang surface growth in (2+1) dimensions. Key physical insights about this universality class, like precise universal exponent values and exponent relations, obtained from large-scale simulations are presented.
At the end of the talk, I will also speak about my current work at the computational science group at HZDR, which includes problems like frameworkdevelopment, image analysis and related machine learning applications.
Keywords: Lattice Monte Carlo, GPU, Surface Growth, Kardar-Parisi-Zhang
  • Lecture (Conference)
    IHRS NanoNet Annual Workshop 2017, 16.-18.08.2017, Neuklingenberg, Deutschland

Registration No. 25961 - Permalink


GPU-Accelerated Kinetic Lattice Monte Carlo for Experimental-Scale Studies
Kelling, J.; Heinig, K. H.; Weigel, M.; Gemming, S.
Micro- and nano-structured materials, including composites, are crucial for future energy technologies. Key processes during production and life-time are governed by self-organization in phase separation processes at the micro and nano scale.
Examples include nano-structured Silicon thin film absorber layers in solar cells providing tailored band-gaps [1] on top of reduced production cost. In the case of micro-patterned electrolyte-matrices, used in a range of fuel cell technologies, both production and aging are governed by phase separation and affect the efficiency and lifetime of large industrial installations.

Simulations of these out-of-equilibrium, inhomogeneous real world systems provide important insights, finding reaction pathways for self-organization and self-alignment of nanostructures. To this end, 3D kinetic Metropolis lattice Monte Carlo simulations can be used to model physical systems at experimental scales in an atomistic way, thereby side-stepping many caveats connected with the alternative phase-field simulations.

These same long-time and large-scale simulations also provide important insights into more fundamental physical problems. The question of super-universality, that is if and how different types of quenched disorder affect universal properties, is still under investigation even for fundamental models like Ising [2], realization of which can be found in important complex magnetic systems, apart from binary mixtures.

We propose massively parallel simulation techniques using the architecture of modern graphics processing units (GPUs) to address these problems, ranging from the kinetic Metropolis Monte Carlo to any Potts models with quenched disorder.
While pioneering work in this area [3] focused on efficient but correlated stochastic cellular automaton implementations, our simulations can be virtually correlation-free [4].

Here, we present two implementations for large-scale simulations on GPUs: One is optimized to offer fast time-to-solution on experimental-scale simulations [5], the other provides highly efficient parameter studies or large sample sizes for large-scale simulations [6]. Harnessing the compute power of modern (multi-)GPU installations leads to increased energy efficiency as well as reduced time-to-solution.

[1] Apl. Phys. Lett. 103, 133106 (2013); Appl. Phys. Lett. 103, 203103 (2013); Nanolett. 16, 1942 (2016)
[2] e.g. EPL 117, 10012 (2017); Phys. Rev. B 88 042129 (2013); Phys. Rev. B 78 224419 (2012); J. Phys. A 24 L1087 (1991)
[3] J. Comp. Phys. 228, 4469 (2009); Phys. Proc. 15 92 (2001)
[4] http://arxiv.org/abs/1705.01022 ; https://arxiv.org/abs/1701.03638
[5] EPJST 210, 175 (2012)
[6] Phys. Rev. E 94 022107 (2016)
Keywords: Kinetic Lattice Monte-Carlo, GPU, Non-equilibrium, Selforganization
  • Lecture (others)
    TYC@Imperial, 16.10.2017, London, England
  • Lecture (Conference)
    2017 MRS Fall Meeting, 26.11.-01.12.2017, Boston, USA

Registration No. 25899 - Permalink


OpenACC for Programmers - Concepts and Strategies
Chandrasekaran, S.ORC; Juckeland, G.ORC
Klappentext noch nicht vorhanden
  • Book (Editorship)
    New York, USA: Addison-Wesley Professional, 2017
    320 Seiten
    ISBN: 978-0134694283

Registration No. 25672 - Permalink


Analyzing Offloading Inefficiencies in Scalable Heterogeneous Applications
Dietrich, R.ORC; Tschüter, R.; Juckeland, G.ORC; Knüpfer, A.
With the rise of accelerators in high performance computing, programming models for the development of heterogeneous applications have evolved and are continuously being improved to increase program performance and programmer productivity. The concept of computation offloading to massively parallel compute devices has established itself as a new layer of parallelism in scientific applications, next to message passing and multi-threading. To optimize the execution of a respective parallel heterogeneous program for a specific platform, performance analysis is crucial. This work abstracts from specific offloading APIs such as available with CUDA, OpenCL, OpenACC, and OpenMP and summarizes common inefficiencies for offloading. Based on the definition of inefficiency patterns, the offloading concept can be included in generic analysis techniques such as critical-path and root-cause analysis. We implemented the detection and evaluation of inefficiency patterns as a post-mortem trace analysis, which finally highlights program activities with a high potential to reduce the total program runtime.
Keywords: performance analysis computation offloading heterogeneous applications critical path
  • Contribution to proceedings
    2nd International Workshop on Performance Portable Programming Models for Accelerators (P^3MA), colocated with the ISC High Performance Conference in Frankfurt, Germany, 22.06.2017, Frankfurt/Main, Deutschland
    High Performance Computing. ISC High Performance 2017. Lecture Notes in Computer Science, vol 10524, Cham: Springer, 978-3-319-67630-2, 457-476
    DOI: 10.1007/978-3-319-67630-2_34

Registration No. 25671 - Permalink


Efficient Parallel Monte-Carlo Simulations for Large-Scale Studies of Surface Growth Processes
Kelling, J.
Lattice Monte Carlo methods are used to investigate far from and out-of-equilibrium systems, including surface growth, spin systems and solid mixtures. Such studies require observations of large systems over long times scales, to allow structures to grow over orders of magnitude, which necessitates massively parallel simulations. This talk presents work done to address the problem of parallel processing introducing correlations in Monte Carlo updates. Studies of the effect of correlations on scaling and dynamical properties of surface growth systems and related lattice gases is investigated further by comparing results obtained by correlation-free and intrinsically correlated but highly efficient simulations using a stochastic cellular automaton. The primary subject of study is the Kardar-Parisi-Zhang surface growth in (2+1) dimensions. Key physical insights about this universality class, like precise universal exponent values and exponent relations, obtained from large-scale simulations are presented.
Keywords: Lattice Monte Carlo, GPU, Surface Growth, Kardar-Parisi-Zhang
  • Lecture (others)
    Seminar Topical Problems, 14.06.2017, Chemnitz, Deutschland

Registration No. 25653 - Permalink


Efficient Correlation-Free Many-States Lattice Monte Carlo on GPUs
Kelling, J.; Ódor, G.; Weigel, M.; Gemming, S.
We'll present a method for highly efficient lattice Monte Carlo simulations with correlation-free updates. Achieving freedom from erroneous correlations requires random selection of lattice sites for updates, which must be restricted by suitable domain decomposition to create parallelism. While approaches based on caching limit the number of allowed states, the multisurface-type approach presented here allows arbitrarily complex states. The effectiveness of the method is illustrated in the fact that it allowed us to solve a long-standing dispute around surface growth under random kinetic deposition in the KPZ-universality class. The method has also been applied to Potts models and is suitable for spin-glass simulations, such as those required to test quantum annealers, like D-Wave.
Keywords: Lattice Monte Carlo, GPU, Surface Growth, Kardar-Parisi-Zhang

Registration No. 25585 - Permalink


Classification at 10Hz: Protecting High-Power Lasers with Deep Learning
Kelling, J.; Gebhardt, R.; Helbig, U.; Bock, S.; Schramm, U.; Juckeland, G.
In this talk we present our approach to automatic detection of critical failure states in pulsed Petawatt laser systems, used for investigations of exotic states of matter and medical applications. The beam shape is controlled to avoid high destructive energy densities. However, randomly occurring states threatening the device must be detected between pulses and trigger an interlock in the device firing at 10Hz.

Our automation approach, presented here, uses deep learning via the Caffe framework. The states we are aiming to detect are rare; thus, training data for this category is scarce. We address this by identifying regions of interest based on physical properties of the system.
Keywords: image classification, deep learning, smart laser operation, OpenCV
  • Lecture (Conference)
    Minds Mastering Machines [M³], 09.-11.10.2017, London, United Kingdom

Registration No. 25584 - Permalink


Suppressing correlations in massively parallel simulations of lattice models
Kelling, J.; Ódor, G.; Gemming, S.
Corresponding author: Kelling, J.
For lattice Monte Carlo simulations parallelization is crucial to make studies of large systems and long Simulation time feasible, while sequential simulations remain the gold-standard for correlation-free dynamics. Here, various domain decomposition schemes are compared, concluding with one which delivers virtually correlation-free simulations on GPUs.
Extensive simulations of the octahedron model for 2 + 1 dimensional Kardar–Parisi–Zhang surface growth, which is very sensitive to correlation in the site-selection dynamics, were performed to show self-consistency of the parallel runs and agreement with the sequential algorithm. We present a GPU implementation providing a speedup of about 30× over a parallel CPU implementation on a single socket and at least 180× with respect to the sequential reference.
Keywords: Lattice Monte Carlo, Kardar-Parisi-Zhang, GPU, autocorrelation

Downloads:

  • Secondary publication expected from 27.07.2018

Registration No. 25479 - Permalink


Profiling of RT-PICLS Code
Kelling, J.; Juckeland, G.
It was observed, that the RT-PICLS code ran by FWKT on the hypnos cluster was producing an unusual amount of system load, according to Ganglia metrics. Since this may point to an IO-problem in the code, this code was analyzed more closely.
Keywords: particle-in-cell, profiling
  • Other report
    Dresden: Qucosa, 2017

Downloads:

Registration No. 25474 - Permalink


Open Access meets Saxony!
Reschke, E.ORC; Stöhr, M.; Kühle, G.
Corresponding author: Reschke, E.
Die Open Access Tage 2017 werden in Dresden unter dem Dach von Dresden concept e.V. gemeinsam vom HZDR, der SLUB Dresden und der TU Dresden organisiert. Der Beitrag stellt die Veranstaltung vor.
Keywords: Open Access
  • Open Access LogoBIS : das Magazin der Bibliotheken in Sachsen 1(2017), 32-33

Registration No. 25175 - Permalink


Dynamical universality classes of simple growth and lattice gas models
Kelling, J.; Odor, G.; Gemming, S.
Corresponding author: Kelling, Jeffrey
Large scale, dynamical simulations have been performed for the two dimensional octahedron model, describing Kardar-Parisi-Zhang (KPZ) for nonlinear, or Edwards-Wilkinson (EW) for linear surface growth. The autocorrelation functions of the heights and the dimer lattice gas variables are determined with high precision. Parallel random-sequential (RS) and two-sub-lattice stochastic dynamics (SCA) have been compared. The latter causes a constant correlation in the long time limit, but after subtracting it one can find the same height functions as in case of RS. On the other hand the ordered update alters the dynamics of the lattice gas variables, by increasing (decreasing) the memory effects for nonlinear (linear) models with respect to RS. Additionally, we support the KPZ ansatz in 2+1 dimensions and provide a precise growth exponent value β = 0.2414(2).. We show the emergence of finite size corrections, which occur much earlier than the height saturation.
Keywords: Non-Equilibrium Thermodynamics, Surface Growth, GPU, Kardar-Parisi-Zhang

Downloads:

  • Secondary publication expected from 14.12.2018

Registration No. 24802 - Permalink


Rapid Data Processing for Ultrafast X-Ray Computed Tomography Using Scalable and Modular CUDA based Pipelines
Frust, T.; Wagner, M.; Stephan, J.; Juckeland, G.; Bieberle, A.
Corresponding author: Frust, Tobias
Ultrafast X-ray tomography is an advanced imaging technique for the study of dynamic processes basing on the principles of electron beam scanning. A typical application case for this technique is e.g. the study of multiphase flows, that is, flows of mixtures of substances such as gas-liquid flows in pipelines or chemical reactors. At Helmholtz-Zentrum Dresden-Rossendorf (HZDR) a number of such tomography scanners are operated. Currently, there are two main points limiting their application in some fields. First, after each CT scan sequence the data of the radiation detector must be downloaded from the scanner to a data processing machine. Second, the current data processing is comparably time-consuming compared to the CT scan sequence interval. To enable online observations or use this technique to control actuators in real-time, a modular and scalable data processing tool has been developed, consisting of user-definable stages working independently together in a so called data processing pipeline, that keeps up with the CT scanner's maximal frame rate of up to 8 kHz. The data processing stages are arbitrarily programmable and combinable. In order to achieve the highest processing performance all relevant data processing steps, which are required for a standard slice image reconstruction, were individually implemented in separate stages using Graphics Processing Units (GPUs) and NVIDIA's CUDA programming language. Data processing performance tests on different high-end GPUs (Tesla K20c, GeForce GTX 1080) and external computer clusters (Tesla P100) showed excellent performance.
Keywords: Computed tomography; Image reconstruction; Multithreading; Parallel algorithms; Pipeline processing; Real-time systems

Downloads:

  • Secondary publication expected from 15.06.2018

Registration No. 24546 - Permalink


Entwicklung und Umsetzung eines echtzeitfähigen Datenverarbeitungs- und Rekonstruktionsalgorithmus für die ultraschnelle Elektronenstrahl-Röntgen-CT
Frust, T.
This thesis demonstrates the development and the implementation of a real-time capable data processing and reconstruction algorithm for the ulftrafast X-ray scanner ROssendorf Fast Electron beam X-ray tomograph (ROFEX). This measuring system is built for non-invasive imaging of multiphase fluids. Thus, it requires a high scan rate of more than 1 kHz. This is achieved by an arrangement without mechanically rotating parts providing scan rates of up to 8 kHz. Current data processing is not suited to reconstruct a data stream of around 1,3 GB/s. Hence, visual inspection or active process feedback control is not possible yet. This work presents the design and implementation of real-time capable data processing providing a better usability and new fields of application for ROFEX. Therefore, the adjustment of data transfer from the measuring system to the reconstruction workstation as well as the implementation of a new software is essential. The application is implemented using a generic software pipeline consisting of distinct processing units exploiting data parallelism of NVIDIA GPUs with CUDA. Measurements on a NVIDIA GeForce GTX 1080 and NVIDIA Tesla K20c yield a reconstruction rate of more than 1 kHz, which is well-suited for an online application of ROFEX. Furthermore, a test system is introduced simulating an online capable detector. It shows the online visualization of reconstructed images as a first new application.
Keywords: cuda, parallel data processing, in-situ, ROFEX
  • Diploma thesis
    TU Dresden, 2016
    Mentor: Prof. Uwe Hampel
    81 Seiten
  • Lecture (others)
    Verteidigung der Diplomarbeit, 03.11.2016, Dresden, Deutschland

Registration No. 24290 - Permalink


In-Situ Analysis and Experiment Regulation at HZDR
Frust, T.; Kelling, J.; Juckeland, G.
A number of experiments at HZDR require in-situ evaluation of the experiment data of the results for automated responses by the experiment control systems. This talk presents two such examples and their current implementations.
  • Invited lecture (Conferences)
    PADC Annual Workshop 2016, 17.-18.10.2016, Jülich, Deutschland

Registration No. 24222 - Permalink


PIConGPU: First Experiences on Minski
Widera, R.; Hübl, A.; Juckeland, G.
PICOnGPU is a highly scalable GPU accelerated particle-in-cell code. This talk presents the first experiences on running this code on IBM Minski (Power8+ with NVIDIA P100 GPUs).
  • Invited lecture (Conferences)
    PADC Annual Workshop 2016, 17.-18.10.2016, Jülich, Deutschland

Registration No. 24221 - Permalink


In-Depth Performance Analysis for OpenACC/CUDA/OpenCL Applications with Score-P and Vampir
Juckeland, G.; Henschel, R.
Participants will work with Score-P/Vampir to learn how to dive into the execution properties of CUDA and OpenACC applications. We'll show how to use Score-P to generate a trace file and how to study it with Vampir. Additionally, we'll use the newly established OpenACC tools interface to also present how OpenACC applications can be studied for performance bottlenecks. This lab utilizes GPU resources in the cloud, you are required to bring your own laptop.
Keywords: Score-P Vampir Tracing Performance-Analysis
  • Lecture (Conference)
    GTC Europe 2016, 28.-29.09.2016, Amsterdam, Nederland

Registration No. 24143 - Permalink


Local Scale-Invariance of the 2+1 dimensional Kardar-Parisi-Zhang model
Kelling, J.; Ódor, G.; Gemming, S.
Local Scale-Invariance theory is tested by extensive dynamical simulations of the driven dimer lattice gas model, describing the surface growth of the 2+1 dimensional Kardar–Parisi–Zhang surfaces. Very precise measurements of the universal autoresponse function enabled us to perform nonlinear fitting with the scaling forms, suggested by local scale-invariance (LSI). While the simple LSI ansatz does not seem to work, forms based on logarithmic extension of LSI provide satisfactory description of the full (measured) time evolution of the autoresponse function.
Keywords: Kardar-Parisi-Zhang, Local Scale-Invariance, Monte-Carlo, GPU

Downloads:

  • Secondary publication expected from 20.02.2018

Registration No. 24116 - Permalink


Multiscale modeling of bacterial colonies: How pili mediate the dynamics of single cells and cellular aggregates
Pönisch, W.; Weber, C.; Juckeland, G.; Biais, N.; Zaburdaev, V.
Neisseria gonorrhoeae is the causative agent of one of the most common sexually transmitted diseases, gonorrhea. An vital step during the infection process is the formation of microcolonies, agglomerates of up to thousands of cells. The assembly of these colonies is driven by type IV pili, filamentaeous polymers protruding from the surface of the colonies, undergoing cycles of elongation and retraction and forming bonds with each other and a substrate. Here, we present a computer model of individual cells interacting via each other solely by pili. This model allows us to study a wide range of processes, from the motion of individual cells and colonies on a surface, over the dynamics within colonies and how they drive the coalescence of two microcolonies, up to the self-assembly of cells.

Registration No. 24105 - Permalink


WissKom2016: Der Schritt zurück als Schritt nach vorn – Macht der Siegeszug des Open Access Bibliotheken arbeitslos? 7. Konferenz der Zentralbibliothek, Forschungszentrum Jülich
Reschke, E.
Vom 14. bis 16. Juni 2016 fand die nunmehr siebente WissKom statt. Die Zentralbibliothek des Forschungszentrum Jülich hatte wieder eingeladen und in gewohnter Perfektion die Tagung organisiert. An drei Tagen wurde zum Themenfeld Modernes Bibliothekswesen und zeitgemäße Informationsversorgung unter besonderem Augenmerk auf den Open Access Transformationsprozess im wissenschaftlichen Publikationswesen vorgetragen und diskutiert. Neue Aufgabenfelder für wissenschaftliche Bibliotheken im vielgestaltigen Prozess der Neuorientierung wurden vorgestellt. Das umfasste den Wandel im Publikationsprozess selbst, Nachweissysteme für Publikationen, Repositorien, Open Access Grün und Zweitveröffentlichung, Forschungsdaten und Fragen der Wissenschaftsevaluierung.
Keywords: Wisskom, Open Access, Libray
  • b.i.t.online 19(2016)5, 459-460
  • Open Access Logob.i.t.online 19(2016)5, 458-466

Registration No. 24046 - Permalink


Scalable and Modular Online Data Processing for Ultrafast Computed Tomography Using CUDA Pipelines
Frust, T.; Juckeland, G.; Bieberle, A.
For investigations of rapidly moving structures in opaque technical devices ultrafast electron beam X-ray computed tomography (CT) scanners are available at the Helmholtz-Zentrum Dresden-Rossendorf (HZDR). Currently, CT data must be downloaded initially after each CT scan from the scanner to a data processing machine. Afterwards, cross-sectional images are reconstructed.
This limits the application fields of the scanners. For on-line observations and even automated process control of scanned objects a new modular data processing tool is presented consisting of user-definable pipeline stages that work independently together in a so called data processing pipeline that can keep up with the frame rate of up to 8 kHz. The stages are arbitrarily programmable and combinable and are connected by a fast custom memory pool to optimize data transfer processes. As a result, this processing structure is not limited to CT application only. In order to achieve highest processing performances for the electron beam CT scanners all relevant data processing steps are individually implemented in separate stages using graphic processing units (GPUs) and NVIDIA's CUDA programming language. Data processing performance tests on two different high-end GPUs (Tesla K20c, GeForce GTX 1080) offer a slice image reconstruction performance that is well-suited for the required on-line application.
Keywords: Electron beam X-ray computed tomography, CUDA, data pipeline, real-time processing, in-situ visualization
  • Contribution to proceedings
    ISAV 2016: In Situ Infrastructures for Enabling Extreme-scale Analysis and Visualization, 13.11.2016, Salt Lake City, Utah, USA
    Proceedings of ISAV 2016
    DOI: 10.1109/ISAV.2016.007
  • Lecture (Conference)
    ISAV 2016: In Situ Infrastructures for Enabling Extreme-scale Analysis and Visualization, 13.11.2016, Salt Lake City, Utah, USA

Registration No. 24022 - Permalink


Scalable critical-path analysis and optimization guidance for hybrid MPI-CUDA applications
Schmitt, F.; Dietrich, R.; Juckeland, G.
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.
Keywords: GPGPU, CUDA, MPI, wait states, critical-path analysis, performance analysis, performance optimization
  • International Journal of High Performance Computing Applications (2017)
    Online First (2016) DOI: 10.1177/1094342016661865

Registration No. 23968 - Permalink


Pushing the Limits of Lattice Monte-Carlo Simulations using GPUs
Kelling, J.; Ódor, G.; Heinig, K. H.; Weigel, M.; Gemming, S.
Lattice Monte-Carlo methods are used to study out-of- and towards-equilibrium systems, like surface growth, spin systems and even phase separation in solid mixtures using kinetic Metropolis lattice Monte-Carlo (KLMC). Applications range from the study of universal scaling or aging behaviors to concrete systems, where coarsening of nanocomposites or self-organization of functional nanostructures is relevant, for example spinodal decomposition in solar cell absorber layers. In these systems, scaling needs to be followed for long times to allow structures to grow over orders of magnitude, which requires large-scale simulations. For the evolution of nanostructures, atomistic simulations at experimental spatiotemporal scales are often desired.

This talk will give an overview over a variety of lattice Monte-Carlo algorithms, which have been found or made suitable for implementation on GPUs: Stochastic cellular automata can be implemented very efficiently [1-3] and are suitable for many systems. The efficient implementation of random sequential dynamics is more challenging. Solutions will be presented for a dimer lattice gas mapped to surface growth [4,5] and KLMC [6]. The latter was also extended to implement dynamics driven by ion-beam mixing triggering long-range interactions. However, these implementations hinge on the fact, that only a very small number of states need to be encoded at each lattice site. A more flexible implementation, employing a variation of multisurface-coding to enable vectorization, will be presented for simulations of restricted solid-on-solid and Potts models with random sequential dynamics. [7]

[1] Block, B., Virnau, P., Preis, T.: Comp. Phys. Comm. 181(9), 1549 (2010)
[2] Lulli, M., Bernaschi, M., Parisi, G.: Comp. Phys. Comm. 196, 290 (2015)
[3] Kelling, J., Ódor, G., Gemming, S.: 2016 IEEE Int. Conf. Intell. Eng. Syst., arXiv:1606.00310 (2016)
[4] Kelling, J., Ódor, G.: Phys. Rev. E 84, 061150 (2011)
[5] Ódor, G., Kelling, J., Gemming, S.: Phys. Rev. E 89, 032146 (2014)
[6] Kelling, J., Ódor, G., Nagy, M. F., Schulz, H., Heinig, K.: EPJST 210, 175 (2012)
[7] Kelling, J., Ódor, G., Gemming, S.: arXiv:1605.02620 (2016)
  • Invited lecture (Conferences)
    Perspectives of GPU computing in Science, 26.-28.09.2016, Roma, Italia

Registration No. 23942 - Permalink


Experimental-Scale Kinetic Lattice Monte-Carlo Studies on GPU
Kelling, J.; Heinig, K.-H.; Gemming, S.
Resume : Micro- and nano-structured materials are crucial for future energy technologies. Key processes during production and life-time are governed by self-organization in phase separation processes at the micro and nano scale. Examples include nano-structured Silicon absorber layers in solar cells providing tailored band-gaps [Apl. Phys. Lett. 103, 133106 (2013)] as well as cheap production and micro-patterned electrolyte-matrices] enhancing life-time and efficiency in a range of fuel cell technologies. Simulations of these out-of-equilibrium, inhomogeneous real world systems provide important insights, finding potential for optimization of structures and process parameters. To this end, kinetic lattice Monte Carlo simulations can be used to model physical systems at experimental scales in an atomistic way, thereby side-stepping many caveats connected with the alternative phase-field simulations. In this contribution, we present two massively parallel implementations for large-scale simulations on GPUs: One is optimized to offer fast time-to-solution on experimental-scale simulations [Eur. J. Phys.: Spec. Top. 210, 175 (2012)], the other provides highly efficient parameter studies or large sample sizes for large-scale simulations [Phys. Rev. E (2016) submitted]. Harnessing the compute power of modern (multi-)GPU installations leads to increased energy efficiency as well as reduced time-to-solution.
  • Lecture (Conference)
    E-MRS Fall Meeting, 19.-22.09.2016, Warschau, Polen

Registration No. 23941 - Permalink


Aging In The (2+1)-Dimensional Kardar-Parisi-Zhang Model Under Various Dimer Lattice-Gas Dynamics
Kelling, J.; Ódor, G.; Gemming, S.
Extensive dynamical simulations of a 2 dimensional driven dimer lattice gas are presented, which can be mapped to (2+1) dimensional surface growth in the Kardar-Parisi-Zhang (KPZ) or Edwards-Wilkinson universality classes [1,2]. From this, autocorrelation and autoresponse functions have been determined for the KPZ universality class and the underlying lattice gas [3]. Studying the effects of different dimer lattice gas dynamics revealed strong differences in the aging behavior of the stochastic cellular automaton (SCA) and the random sequential update models. We show numerical evidence for nontrivial corrections and tests against log-local scale invariance [4] as well as different universal scaling behaviors. [1] G. Ódor, B. Liedke and K.-H. Heinig, Phys. Rev. E 79, 021125 (2009) [2] J. Kelling and G. Ódor, Phys. Rev. E 84, 061150 (2011) [3] J. Kelling, G. Ódor, S. Gemming, Phys. Rev. E 89, 032146 (2014) [4] M. Henkel, Nucl. Phy. B 869(2), 282 (2013)
  • Lecture (Conference)
    Stat'Phys 26 - Statistical Physics Conference Satellite Non-equilibrium dynamics in classical and quantum systems: From quenches to slow relaxations, 13.-22.07.2016, Pont-à-Mousson, France

Registration No. 23940 - Permalink


Profiling Performance of hybrid applications with Score-P and Vampir
Juckeland, G.; Dietrich, R.
OpenACC aims at providing a relatively easy and straightforward way to describe parallelism for exploitation on platforms with hardware accelerators. It is by design also an approach for porting legacy HPC applications to this novel architecture. Especially such legacy applications, but also newly developed applications that require more resources than a single node can offer, use MPI for inter-node communication and coarse work distribution, thus, becoming so-called hybrid applications. It is also possible to combine OpenACC with OpenMP on the host side to utilize all resources of a compute node or to even use all three levels of parallelism concurrently. Tuning application performance for one parallelization paradigm is challenging, adding the second or third level of parallelism introduces a whole new layer of potential performance problems from the interaction of all parallelization paradigms. It is, however, possible to extend the previously mentioned profile-guided development to also cover this usage scenario.
Profiling tools from compiler or accelerator vendors are usually limited to the scheme the product address, e.g. only OpenACC or CUDA/OpenCL activity. Almost all vendor tools cannot record MPI activity leaving the programmer in the dark how well hybrid applications perform over all used levels of parallelism. Research based performance tools cover this gap. HPCtoolkit, Tau, and Score-P are the most prominent ones that also offer hardware accelerator support. Out of the three Score-P is the one that covers the most parallelization paradigms, can record the most concurrent activity and, as a result, can provide the most complete performance picture even for very complex applications. Therefore, Score-P will be used as the example performance recording tool for this chapter. The other tools can provide similar results. Vampir will be use for visualizing the performance data since it is by far the most capable trace visualizer and profile generator.
  • Book chapter
    Farber, Rob: OpenACC - Parallel Programming with OpenACC, Amsterdam: Elsevier, 2016, 978-0-12-410397-9, 55-68

Registration No. 23815 - Permalink


Bit-Vectorized GPU Implementation of a Stochastic Cellular Automaton Model for Surface Growth
Kelling, J.; Ódor, G.; Gemming, S.
Stochastic surface growth models aid in studying properties of universality classes like the Kardar--Parisi--Zhang class. High precision results obtained from large scale computational studies can be transferred to many physical systems. Many properties, such as roughening and some two-time functions can be studied using stochastic cellular automaton (SCA) variants of stochastic models. Here we present a highly efficient SCA implementation of a surface growth model capable of simulating billions of lattice sites on a single GPU. We also provide insight into cases requiring arbitrary random probabilities which are not accessible through bit-vectorization.
Keywords: Surface Growth, Kardar-Parisi-Zhang, GPU, Monte-Carlo, Stochastic Cellular Automaton
  • Contribution to proceedings
    20th Jubilee IEEE International Conference on Intelligent Engineering Systems 2016, 30.06.-02.07.2016, Budapest, Ungarn
    Proceedings of the 20th Jubilee IEEE International Conference on Intelligent Engineering Systems 2016: IEEE
    DOI: 10.1109/INES.2016.7555127
  • Lecture (Conference)
    20th Jubilee IEEE International Conference on Intelligent Engineering Systems 2016: IEEE, 30.06.-02.07.2016, Budapest, Ungarn
  • Contribution to WWW
    https://arxiv.org/abs/1606.00310

Registration No. 23689 - Permalink


From Describing to Prescribing Parallelism: Translating the SPEC ACCEL OpenACC Suite to OpenMP Target Directives
Juckeland, G.; Hernandez, O.; Jacob, A. C.; Neilson, D.; Vergara Larrea, V. G.; Wienke, S.; Chandrasekaran, S.; Grund, A.; Henschel, R.; Müller, M. S.; Perminov, M.; Shelepugin, P.; Whitney, B.; Joubert, W.; Wang, B.; Kumaran, K.; SPEC High Performance Group (HPG), Email: info@spec.org – Website: http://www.spec.org/hpg; Bobyr, A.
Current and next generation HPC systems will exploit accelerators and self-hosting devices within their compute nodes to accelerate applications. This comes at a time when programmer productivity and the ability to produce portable code has been recognized as a major concern. One of the goals of OpenMP and OpenACC is to allow the user to specify parallelism via directives so that compilers can generate device specific code and optimizations. However, the challenge of porting codes becomes more complex because of the different types of parallelism and memory hierarchies available on different architectures.
In this paper we discuss our experience with porting the SPEC ACCEL benchmarks from OpenACC to OpenMP 4.5 using a performance portable style that lets the compiler make platform-specific decisions to achieve performance. The SPEC ACCEL OpenMP benchmarks were tested on different platforms including Xeon Phi, GPUs and CPU cores. We also discuss some performance-portable best practices that help expose more parallelism in the codes in a portable style. We believe that this expe rience can help the community and compiler vendors understand how users plan to write OpenMP 4.5 applications in a performance portable
style.
Keywords: SPEC, SPEC ACCEL, OpenMP, OpenACC, Offloading
  • Contribution to proceedings
    Workshop on Performance Portable Programming Models for Accelerators (P^3MA), 23.06.2016, Frankfurt (Main), Deutschland
    Lecture Notes on Computer Science (LNCS) - Volume 9934: High Performance Computing ISC High Performance 2016 International Workshops, ExaComm, E-MuCoCoS, HPC-IODC, IXPUG, IWOPH, P^3MA, VHPC, WOPSSS, Frankfurt, Germany, June 19–23, 2016: Springer, 978-3-319-46078-9, 470-488
    DOI: 10.1007/978-3-319-46079-6_33

Registration No. 23664 - Permalink


Universality of (2+1)-dimensional restricted solid-on-solid models
Kelling, J.; Ódor, G.; Gemming, S.
Extensive dynamical simulations of Restricted Solid on Solid models in D=2+1 dimensions have been done using parallel multisurface algorithms implemented on graphics cards. Numerical evidence is presented that these models exhibit KPZ surface growth scaling, irrespective of the step heights N. We show that by increasing N the corrections to scaling increase, thus smaller step sized models describe better the asymptotic, long wave scaling behavior.

Downloads:

Registration No. 23645 - Permalink


Open Access-Transformationsinitiative "OA2020"
Reschke, E.
Vorstellung der Open Access Transformationsinitiative OA2020
Keywords: Open Access, OA2020, publication fees, author publication charge, White Paper, DEAL
  • Lecture (others)
    Arbeitskreis Spezialbibliotheken Dresden - Frühjahrstreffen,, 20.04.2016, Dresden, Deutschland

Registration No. 23577 - Permalink


Kerndatensatz Forschung
Reschke, E.
Vorstellung der Kategorien, die sich auf das Publizieren beziehen.
Keywords: Kerndatensatz, Research, Publications
  • Lecture (others)
    Arbeitskreis Spezialbibliotheken Dresden - Frühjahrstreffen, 20.04.2016, Dresden, Deutschland

Registration No. 23576 - Permalink


Porting the Plasma Simulation PIConGPU to Heterogeneous Architectures with Alpaka
Zenker, E.; Widera, R.; Juckeland, G.; Worpitz, B.; Hübl, A.; Knüpfer, A.; Nagel, W. E.; Bussmann, M.
We discuss our experience on porting the CUDA-based plasma simulation code PIConGPU to heterogeneous platforms using the abstract kernel interface Alpaka. With the advent of next-generation architectures such as OpenPower, the full use of the hardware and the mapping of CPUs and GPUs to specific simulation tasks has become important. Performance portability is of great interest, but even more important is the ability to develop against a single interface to keep code testable and maintainable. We show how we can make use of the Alpaka library in real-world applications and how we achieve portability and performance.
Keywords: Supercomputing & HPC, Computational Physics, Performance Optimization
  • Lecture (Conference)
    GPU Technology Conference, 04.-07.04.2016, San Jose, California, USA

Registration No. 23569 - Permalink


Performance-Portable Many-Core Plasma Simulations: Porting PIConGPU to OpenPower and Beyond
Zenker, E.; Widera, R.; Hübl, A.; Juckeland, G.; Knüpfer, A.; Nagel, W. E.; Bussmann, M.
With the appearance of the heterogeneous platform OpenPower, many-core accelerator devices have been coupled with Power host processors for the first time. Towards utilizing their full potential, it is worth investigating performance portable algorithms that allow to choose the best-fitting hardware for each domain-specific compute task. Suiting even the high level of parallelism on modern GPGPUs, our presented approach relies heavily on abstract meta-programming techniques, which are essential to focus on fine-grained tuning rather than code porting. With this in mind, the CUDA-based open-source plasma simulation code PIConGPU is currently being abstracted to support the heterogeneous OpenPower platform using our fast porting interface cupla, which wraps the abstract parallel C++11 kernel acceleration library Alpaka.

We demonstrate how PIConGPU can benefit from the tunable kernel execution strategies of the Alpaka library, achieving portability and performance with single-source kernels on conventional CPUs, Power8 CPUs and NVIDIA GPUs.
Keywords: OpenPower, heterogeneous computing, HPC, C++11, CUDA, OpenMP, particle-in-cell, platform portability, performance portability
  • Contribution to proceedings
    International Workshop on OpenPower co-located with ISC High Performance, 23.06.2016, Frankfurt, Germany
    Lecture Notes in Computer Science, Vol. 9945, 293-301
    DOI: 10.1007/978-3-319-46079-6_21

Registration No. 23565 - Permalink


Alpaka - An Abstraction Library for Parallel Kernel Acceleration
Zenker, E.; Worpitz, B.; Widera, R.; Hübl, A.; Juckeland, G.; Knüpfer, A.; Nagel, W. E.; Bussmann, M.
Porting applications to new hardware or programming models is a tedious and error prone process. Every help that eases these burdens is saving developer time that can then be invested into the advancement of the application itself instead of preserving the status-quo on a new platform.

The Alpaka library defines and implements an abstract hierarchical redundant parallelism model. The model exploits parallelism and memory hierarchies on a node at all levels available in current hardware. By doing so, it allows to achieve platform and performance portability across various types of accelerators by ignoring specific unsupported
levels and utilizing only the ones supported on a specific
accelerator. All hardware types (multi- and many-core CPUs, GPUs and other accelerators) are supported for and can be programmed in the same way.
The Alpaka C++ template interface allows for straightforward extension of the library to support other accelerators and specialization of its internals for optimization.

Running Alpaka applications on a new (and supported) platform requires the change of only one source code line instead of a lot of #ifdefs.
Keywords: Heterogeneous computing, HPC, C++, CUDA, OpenMP, platform portability, performance portability
  • Contribution to proceedings
    The Sixth International Workshop on Accelerators and Hybrid Exascale Systems co-located with the 30th IEEE International Parallel & Distributed Processing Symposium, 23.-27.05.2016, Chicago Hyatt Regency Chicago, Illinois, USA
    Proceedings of the 30th IEEE International Parallel and Distributed Processing Symposium

Registration No. 23564 - Permalink


Aging Universality Classes in Surface Growth Models
Kelling, J.; Ódor, G.; Gemming, S.
Extensive dynamical simluations of a 2 dimensional driven dimer lattice gas are presented, which can be mapped to (2+1) dimensional surface growth in the Kardar-Parisi-Zhang (KPZ) or Edwards-Wilkinson unversality classes. From this autocorrelation and autoresponse functions have been determined for the KPZ universality class and the underlying lattice gas. Studying the effects of different dimer lattice gas dynamics revealed strong differences in the aging behavior of the stochastic cellular automaton (SCA) and the random sequential update models. We show numerical evidence for nontrivial corrections as well as different universal scaling behaviors.
Keywords: Surface Growth, Aging, Kardar-Parisi-Zhang, Monte-Carlo, Cellular Automaton
  • Poster
    DPG-Frühjahrstagung der Sektion Kondensierte Materie, 06.-11.03.2016, Regensburg, Deutschland
  • Poster
    VII GEFENOL Summer School on Statistical Physics of Complex Systems, 19.-30.06.2017, Palma de Mallorca, Spain

Registration No. 23300 - Permalink


Non-Local Lattice Encoding for Bit-Vectorized Cellular Automata GPU Implementations
Kelling, J.; Ódor, G.; Gemming, S.
In many areas from physics to economics and social sciences, there are current problems that can be mapped to stochastic cellular automata (SCA). In combination with machine learning techniques, cellular automata with learned rules can be used to efficiently predict real world systems. In physics, they are used to study atomistically the size and shape evolution of micro- and nanostructures, providing insights into processes of self-organization crucial to today's nanotechnology. We present an extremely efficient SCA implementation of a surface growth model using bit-vectorization enhanced by non-local encoding on GPU. The employed technique and non-local encoding can be transfered to other applications.
Keywords: GPGPU, Surface Growth, Kardar-Parisi-Zhang, Monte-Carlo, Cellular Automaton

Registration No. 23299 - Permalink


Aging in the (2+1)-Dimensional Kardar-Parisi-Zhang Model under Various Dimer Lattice-Gas Dynamics
Kelling, J.; Ódor, G.; Gemming, S.
Extensive dynamical simluations of a 2 dimensional driven dimer lattice gas are presented, which can be mapped to (2+1) dimensional surface growth in the Kardar-Parisi-Zhang (KPZ) or Edwards-Wilkinson unversality classes. From this autocorrelation and autoresponse functions have been determined for the KPZ universality class and the underlying lattice gas. Studying the effects of different dimer lattice gas dynamics revealed strong differences in the aging behavior of the stochastic cellular automaton (SCA) and the random sequential update models. We show numerical evidence for nontrivial corrections as well as different universal scaling behaviors.
Keywords: Surface Growth, Aging, Kardar-Parisi-Zhang, Monte-Carlo, Cellular Automaton
  • Poster
    Conference of the Middle European Cooperation in Statistical Physics, 14.-17.02.2016, Wien, Österreich

Registration No. 23298 - Permalink


Elastic and Piezoresistive Properties of Nickel Carbides from First-Principles
Kelling, J.; Zahn, P.; Schuster, J.; Gemming, S.
The nickel-carbon system has received increased attention over the past years due to the relevance of nickel as a catalyst for carbon nanotube and graphene growth, where Nickel carbide intermediates may be involved or carbide interface layers form in the end. Nickel-carbon composite thin films comprising Ni3C are especially interesting in mechanical sensing applications. Due to the meta-stability of nickel carbides, formation conditions and the coupling between mechanical and electrical properties are not yet well understood. Using first-principles electronic structure methods, we calculated the elastic properties of Ni3C, Ni2C and NiC, as well as changes in electronic properties under mechanical strain. We observe that the electronic density of states around the Fermi level does not change under the considered strains of up to 1%, which correspond to stresses up to 3GPa. Relative changes in conductivity of Ni3C range up to maximum values of about 10%.
Keywords: nickel, carbides, thin films, density functional theory, electronic structure, first-principles calculations, transport theory, electrical conductivity, piezoresistivity, Boltzmann theory

Downloads:

Registration No. 23291 - Permalink


Don’t forget the wire
Bussmann, M.; Zacharias, M.; Schöne, S.; Jung, F.; Vogt, S.; Widera, R.; Lieber, S.; Konrad, U.; Kaever, P.; Bieberle, A.; Wagner, A.; Fiedler, F.
Overview of HZDR DTS Activities in Parallel Data Analyis
Keywords: dts, data, analysis, parallel
  • Lecture (Conference)
    Matter and Technologies Kickoff Meeting, 23.-26.02.2015, Hamburg, Deutschland

Registration No. 23226 - Permalink


Alpaka, GrayBat and other spiritual animals that will help you survive in the dangerous world of HPC
Bussmann, M.; Eckert, C.; Huebl, A.; Jung, F.; Widera, R.; Worpitz, B.; Zacharias, M.; Zenker, E.; Juckeland, G.; Knüpfer, A.; Nagel, W.
Implementing codes for simulations or large-scale, high rate data analysis on modern compute hardware can take much of the time that you initially wanted to spend on something entirely else. We present our tool box of simple yet powerful solutions for writing hardware-independent, scalable and performant codes that are maintanable and can survive the next compute hardware update. We show that with modern programming techniques it is possible to implement abstract interfaces that provide sustainable concepts for parallel programming while at the same time allowing for performance optimization. This talk is about real life applications that need to work. But it will also be cool.
Keywords: hpc, c++, simulation, plasma, imaging, abstraction
  • Invited lecture (Conferences)
    ZIH-Kolloquium, 28.01.2016, Dresden, Deutschland

Registration No. 23223 - Permalink


Bibliothekskonzeption als eine Möglichkeit, Konflikte zu vermeiden
Reschke, E.
Bis zum Jahr 2009 gab es immer wieder Diskussionen, ob es mit der Bibliothek am Helmholtz-Zentrum Dresen-Rossendorf weitergehen soll und wenn ja wie, insbesondere hinsichtlich der Personalstärke und des Raumbedarfs (Bibliothek als Ort in einer Forschungseinrichtung).
Daraufhin habe ich eine Bibliothekskonzeption für die Jahre 2010-2014 entwickelt, die zur Entscheidungsgrundlage wurde. Die Konzeption wurde von unserer Bibliothekskommission als richtungsweisend eingeschätzt und gegenüber dem Vorstand des Forschungszentrums zur Umsetzung empfohlen. Jetzt, 2015, können wir die vollständige Umsetzung der Konzeption berichten. Es gibt z.Z. keine Diskussionen um Bestand und Aufgaben der Bibliothek. Zwischen dem Vorstand, der Bibliothekskommission, unserem Zentralabteilungsleiter und der Bibliothek besteht Einvernehmen, dass sich zwar die Aufgaben der Bibliothek in ihrer Wichtung ändern, aber an der Notwendigkeit der Bibliothek vor Ort wird aktuell nicht gezweifelt.
Unterm Strich: Mit der gemeinsame Umsetzung der Konzeption wurde der Konflikt "Bibliothek brauchen wir-brauchen wir nicht" gelöst. Es entstand Planungssicherheit, die natürlich über den Zeitraum der Konzeptionsplanung hinaus weiter verteidigt/erkämpft werden muss.
Keywords: Strategisches Bibliotheksmanagement, Bibliothekskonzeption
  • Invited lecture (Conferences)
    35. Arbeitstagung der Arbeitsgemeinschaft der Spezialbibliotheken (ASpB), 23.-25.09.2015, Bonn, Deutschland

Registration No. 22525 - Permalink


Die Bibliothek: das Publikations- und Open Access Kompetenzzentrum am HZDR
Reschke, E.
Die Publikationsdatenbank im Helmholtz-Zentrum Dresden-Rossendorf wird seit 1992 geführt. Anfangs wurden nur bibliographische Daten erfasst. Mit den Jahren entstanden Bedürfnisse, mit diesen Daten mehr zu machen als Publikationslisten und es sollte keine doppelte Datenerfassung und Datenvorhaltung geben. Dafür mussten weitere Daten und Parameter in der einen Publikationsdatenbank erfasst werden. Der Genehmigungsworkflow wurde eingebaut, Controllingabfragen mussten möglich gemacht werden, wachsendes Berichtswesen erforderte zunehmende Flexibilität in den Abfragemöglichkeiten, Open Access Finanzierung, Verlinkung zur Datenarchivierung und weiteres. Ergebnis ist eine proprietäre Datenbankanwendung, basierend auf Oracle, die durch ihre Komplexität den Anforderungen aller beteiligten Akteure: Wissenschaftler, Bibliothek, wissenschaftliches Controlling, Institutsdirektoren und Vorstand, gerecht werden kann. Im Vortrag werden die verschiedenen Funktionen und die Organisation der Publikationsdatenbank vorgestellt.
Keywords: Library, Open Access, Publication database, Bibliothek, Publikationsdatenbank
  • Invited lecture (Conferences)
    Fraunhofer Arbeitstagung Bibliotheks- und Informationswesen 2015, 19.-22.10.2015, Fulda, Deutschland

Registration No. 22524 - Permalink


Electronic Transport through Au-contacted PEEB
Kelling, J.; Sendler, T.; Erbe, A.; Gemming, S.
Abstract Transport through the organic molecule 1,4-bis(phenylethynyl)-2,5-bis(ethoxy)benzene (PEEB) has been investigated using Density functional Theory (DFT) and the Non-Equilibrium Green's Function (NEGF) approach in order to explain results of experiments employing the mechanically controlled break junction (MCBJ) technique. The molecule was studied with various terminal groups (NH2, SH and CN) which lead to different conductance values when attached to Gold electrodes. The effect of different contact geometries was studied in simulations, allowing predictions on the most likely contact geometries occurring in experiments.
  • Poster
    NanoNet Workshop, 30.09.-02.10.2015, Rathen, Sächsische Schweiz, Deutschland

Registration No. 22463 - Permalink


Electronic Transport through Au-contacted, Thiol-terminated PEEB
Kelling, J.; Sendler, T.; Erbe, A.; Gemming, S.
Transport through the organic molecule 1,4-bis(phenylethynyl)-2,5-bis(ethoxy)benzene (PEEB) with Thiol (SH) terminal groups, contacted by Gold electrodes, has been investigated using Density functional Theory (DFT) and the Non-Equilibrium Green’s Function (NEGF) method. The results confirm the existence of a single-atom contact configuration with unusually high conductance for organic molecules of >0.1G0.
  • Lecture (others)
    NanoNet Workshop, 30.09.-02.10.2015, Rathen, Sächsische Schweiz, Deutschland

Registration No. 22462 - Permalink


C++11/14 features relevant in GPGPU APIs
Kelling, J.
  • Invited lecture (Conferences)
    GPU Day 2015 - The Future of Many-Core Computing in Science, 20.05.2015, Budapest, Hungary

Registration No. 21995 - Permalink


Efficient Large Scale Simulation of Stochastic Lattice Models on GPUs
Kelling, J.; Ódor, G.; Heinig, K.-H.; Gemming, S.
With growing importance of nano-patterned surfaces and nano-composite materials in many applications from energy technologies to nano-electronics, a thorough understanding of the self-organized evolution of nano-structures needs to be established. Modelling and simulations of such processes can help in this endeavor and provide predictions for the turnout of manufacturing processes.

In this talk GPGPU-enabled implementations of two stochastic lattice models will be discussed, shedding light on the complications which arise when simulations of stochastic processes are to make efficient use of massively parallel GPU architectures.
A single-GPU implementation of the (2+1)-dimensional Roof-Top-model allows very precise large-scale studies of surface growth processes in the Kardar-Parisi-Zhang universality class.[1] Furthermore a multi-GPU enabled version of the 3d kinetic Metropolis lattice Monte-Carlo method[2] provides the capability to study the evolution of nano-structures both towards and out-of-equilibrium at spatio-temporal scales of experiments using only small to medium-sized GPU clusters.

[1] J. Kelling, G. Ódor Extremely large-scale simulation of a Kardar-Parisi-Zhang model using graphics cards, Physical Review E 84, 061150 (2011)
[1] J. Kelling, G. Ódor, F. Nagy, H. Schulz, K. Heinig Comparison of different parallel implementations of the 2+1-dimensional KPZ model and the 3-dimensional KMC model, The European Physical Journal - Special Topics 210, 175-187 (2012)
  • Lecture (others)
    Seminar Topical Problems, 06.05.2015, Chemnitz, Deutschland
  • Invited lecture (Conferences)
    GPU Day 2015 - The Future of Many-Core Computing in Science, 20.-21.05.2015, Budapest, Hungary
  • Lecture (others)
    Seminar, 18.11.2015, Coventry, United Kingdom

Registration No. 21951 - Permalink


Bibliotheksservices für Mitarbeiter des VKTA
Reschke, E.
VKTA-MitarbietrInnen sind externe Bibliotheksbenutzer. Im Vortrag werden die für externe Benutzer verfügbaren Bibliotheks- und Informationsservices vorgestellt.
Keywords: Library, Information services, Research Portal, eJournals, eBooks, Data bases
  • Lecture (others)
    VKTA-Leitungskreissitzung, 28.04.2015, Dresden-Rossendorf, Deutschland

Registration No. 21928 - Permalink


Investigating Spinodal Decomposition and Coarsening using Massively Parallel Kinetic Metropolis Lattice Monte-Carlo Simulations
Kelling, J.; Heinig, K. H.; Gemming, S.
Nano-structured materials are important for many applications, including energy technologies. The desired structures can be created using bottom-up processes, which utilize self-assembling. By way of spinodal decomposition of a metastable phase, like SiOx into Si and SiO2, sponge-like networks of nanowires can be obtained. Understanding the coarsening kinetics of spinodal structures is crucial not only for bottom-up production, but also helps to increase the life-time of components like porous matrices in fuel cells, where suppression of coarsening has a huge economic impact.

Two theories on coarsening of spinodal structures exist: one assuming diffusion through the bulk [1], the other along interfaces [2]. Since orders of magnitude in both space and time have to be covered by simulations, numerical studies are quite demanding. Nevertheless, simulations are essential in studying systems containing size-inhomogeneities in initial nano-structures, where strongly accelerated coarsening is observed.

Here, a multi-GPU Kinetic Metropolis Lattice Monte-Carlo implementation, capable of atomistic simulations of phase-separation and coarsening at spatio-temporal experimental scales (billions of particles over millions of time-steps) is presented while laying focus on the above-mentioned applications.

[1] A. Chakrabarti, R. Toral, J.D. Gunton, Phys. Rev. B 39(7) 4386 (1989) suggesting a modified Lifschitz-Slyozov law: I.M. Lifshitz, V.V. Slyozov, J. Phys. Chem. Solids 19(1-2), 35-50 (1961)
[2] W.W. Mullins, J. Appl. Phys. 28(3), 333-339 (1957)
Keywords: Nano Structures, GPGPU, Supercomputing, Statistical Physics,
  • Lecture (Conference)
    Conference of the Middle European Cooperation in Statistical Physics, 23.-25.03.2015, Esztergom, Hungary

Registration No. 21768 - Permalink


Simulation of Surface Growth and Lattices Gases Using GPUs
Schulz, H.; Kelling, J.; Ódor, G.; Ódor, G.; Ferenc Nagy, M.
Restricted solid on solid surface growth models can be mapped onto binary lattice gases. We show that efficient simulation algorithms can be realized on GPUs either by CUDA or by OpenCL programming. We consider a deposition/evaporation model following Kardar–Parisi–Zhang growth in d+1 dimensions, related to the Asymmetric Simple Exclusion Process. Up to 100 - 400 x speedup can be achieved with respect to the serial code running on a I5 core. This permits studying disorder and aging behavior in these system.
Keywords: Computational Physics, Supercomputing

Registration No. 21767 - Permalink


Handling Domain Decomposition in Massively Parallel Implementations of Stochastic Lattice Models
Kelling, J.; Ódor, G.; Gemming, S.
Nanopatterning of surfaces and bulk materials is very important from molecular electronics to photovoltaics. But, in order to understand the underlying physics of self-organization, large scale atomistic simulations are crucial. Only stochastic models can bridge the gap from nano to micro, enabling simulations of micron-sized volumes, billions of atoms and study long-time evolution. Random site-selection is essential but can be harmed by domain decomposition in GPGPU. We present solutions by example of a dimer-model for KPZ surface growth.
Keywords: Statistical Physics, Domain Decompostion

Registration No. 21484 - Permalink


Transport Calculations for Si4 Clusters with Gold single Atom Contacts
Kelling, J.; Kerbusch, J.; Erbe, A.; Dietsche, R.; Ganteför, G.; Scheer, E.; Zahn, P.; Gemming, S.
We present results for electronic transport through Si4 clusters contacted by single-atom gold contacts attached to <111> fcc-gold leads. The calculations were performed using density functional theory and the non-equilibrium Green's function approach for transport. The simulation setup mimics contacts in mechanically controllable break-junction experiments, which provide data for comparison.
  • Poster
    DPG-Frühjahrstagung der Sektion Kondensierte Materie, 15.-20.03.2015, Berlin, Deutschland

Registration No. 21481 - Permalink


Electronic Transport through Au-contacted PEEB
Kelling, J.; Gemming, S.
Transport through the organic molecule PEEB has been investigated using Density functional Theory (DFT) and the Non-Equilibrium Green's Function (NEGF) approach.
Keywords: electronic transport, molectular electronics, dft
  • Poster
    NanoNet Workshop, 29.-30.09.2014, Rathen, Sächsische Schweiz, Deutschland

Registration No. 21478 - Permalink


Primo – Projekterfahrungen im Helmholtz-Zentrum Dresden-Rossendorf
Reschke, E.
Primo ist eines der am meisten eingesetzten Discovery-Systeme weltweit. Im Helmholtz-Zentrum Dresden-Rossendorf entschied man sich für den Einsatz von Primo Total Care (https://www.hzdr.de/primo) Im Bericht werden die Projekterfahrungen dargestellt, von der Entscheidungsfindung bis zur Akzeptanz durch die Nutzer. Insbesondere werden die erfahrenen Vor- und Nachteile der Total Care-Installation dargestellt.
Keywords: Discovery system, Primo Total Care
  • Invited lecture (Conferences)
    Jahrestagung des AK Bibliotheken der Leibnitz-Gemeinschaft, 10.-12.09.2014, Bonn, Deutschland

Registration No. 20774 - Permalink


Die Bibliothek als Open Access-Kompetenzzentrum in einer außeruniversitären Forschungseinrichtung
Reschke, E.
Open Access zu publizieren, ist noch nicht selbstverständlich für WissenschaftlerInnen in den außeruniversitären Forschungsgemeinschaften. Neue Publikationsmöglichkeiten erfordern neue, prozessbegleitende Services. Der Aufbau eines Open Access Kompetenzzentrums in der Bibliothek ist ein bereits erfolgreicher Weg, die WissenschaftlerInnen umfassend zu informieren und zu unterstützen. Welche Leistungen erbringt das Kompetenzzentrum und wie erwirbt es die notwendige Kompetenz? Im Fallbeispiel 2 soll dargestellt werden, welche administrativen Bereiche das Thema Open Access tangiert, wie diese Bereiche zusammenarbeiten und wie sich dieses Netz für seine Aufgaben qualifiziert.
Keywords: Open Access, Library management,
  • Invited lecture (Conferences)
    Open-Access-Tage 2014, 29.-30.09.2014, Köln, Deutschland

Registration No. 20771 - Permalink


Aging of the (2+1)-dimensional Kardar-Parisi-Zhang model
Ódor, G.; Kelling, J.; Gemming, S.
Extended dynamical simulations have been performed on a (2+1)-dimensional driven dimer lattice-gas model to estimate aging properties. The autocorrelation and the autoresponse functions are determined and the corresponding scaling exponents are tabulated. Since this model can be mapped onto the (2+1)-dimensional Kardar-Parisi-Zhang surface growth model, our results contribute to the understanding of the universality class of that basic system.

Registration No. 20253 - Permalink


Elastic Properties of Nickel Carbides
Kelling, J.; Zahn, P.; Gemming, S.
The nickel -- carbon system has received increased attention over the past years due to the relevance of nickel as a catalyst for carbon nanotube growth and as a substrate for metal-induced crystallization of graphene and graphite.
Nickel carbides as bulk materials are out
of reach for experimental studies because of their meta-stability. Ab-initio studies are rare. We present our ab-initio -- frozen phonon results for the elastic properties of Ni3C, Ni2C and NiC.
  • Poster
    DPG-Frühjahrstagung der Sektion Kondensierte Materie, 30.03.-04.04.2014, Dresden, Germany

Registration No. 20252 - Permalink


GPU-based Atomistic Simulations\\on spatio-temporal experimental Scales
Kelling, J.; Heinig, K.-H.; Gemming, S.
The Kinetic Metropolis Lattice Monte-Carlo (KMC) method is a means of performing atomistic simulations of self-organization processes in solids at by far larger scales than those accessible via Molecular Dynamics. Employing a cellular automaton approach allows incorporation of many body interactions and external driving forces. Here, we present an efficient KMC implementation on single and multiple GPUs, which allows us to study phase separation and nanostructure-evolution at spatio-temporal experimental scales. The KMC implementation has been used to develop with industrial partners a new Si-based nanocomposite for next-generation thin-film solar cells.

Registration No. 20251 - Permalink


Ab-initio investigation of carbides and of CNT junctions at finite temperature and under stress.
Kelling, J.; Gemming, S.
With the presented work we lay the foundation for ab-initio studies of contacted carbon nanotubes with both metal and metal--carbide leads. We Focus on applying the frozen phonon method on top density-functional-theory calculations for electronic the system. Here we show our ab-initio results on the elastic and electronic properties of Al4C3 and the metastable Ni3C.

Furthermore we present an alternative non-perturbative approach to calculating the quantum conductance in CNT/molecular junctions at finite temperature. Strictly employing the Born--Oppenheimer approximation, we aim to calculate the influence of phonons on the conductance of such a system by averaging over a representative sample drawn from snapshots of thermal fluctuation of the lattice.
  • Poster
    NanoNet Workshop, 10.10.2013, Rossendorf, Deutschland

Registration No. 20079 - Permalink


Toward nonconventional human–machine interfaces for supervisory plant process monitoring
Skripcak, T.; Tanuska, P.; Konrad, U.; Schmeisser, N.
This paper describes a novel methodology for designing alternative human–machine interfaces (HMI) dealing with the industrial process control visualization and plant monitoring. The system is based on a multiagent approach in order to allow visualizations using nonconventional display devices (e.g., power-wall or table) combined with the natural user interaction (NUI) paradigm. This type of HMI solution could form an optional extension to current systems employed in a plant monitoring. Namely the utilization of a virtual reality creates new opportunities for the HMI system use cases, where an enhanced visualization and interaction can improve decision making strategies in complex processes. An immersive plant personnel training or realistic process visualizations are examples where the usage of the third dimension can be helpful. Nevertheless, the development of nonconventional HMI increases the complexity of the industrial information system environment. The question is how to adapt the system design for the nonconventional HMI, in a contrast with the conventional solutions. We applied the multiagent approach, which leads to a more robust and less-coupled component structure. An experimental prototype was implemented on the top of the proposed methodology. A simulation of an absorption refrigeration process was used as a process model on top of which the prototype was designed. Interoperability was gained via the automation standard of OPC UA. Furthermore, in order to optimize our NUI agent, a user testing application was developed for the evaluation of exploratory interaction tasks in a power-wall display scenario. The proposed framework provides fundamental guidelines for designing and developing a new generation of HMI systems.
Keywords: Human–machine interfaces; natural user interaction; nonconventional displays; virtual factory; virtual reality

Registration No. 19328 - Permalink


Performing kinetic lattice Monte-Carlo simulations of far-from-equilibrium processes on GPUs
Kelling, J.; Heinig, K.-H.
Kinetic Metropolis lattice Monte-Carlo (KLMC) simulations can be applied to study systems without strong external driving forces, for example the evolution of nanostructures. An algorithm allowing nearest neighbor interaction in thermally activated processes has been implemented for GPUs. External driving forces can be modeled by adding appropriate rules to the cellular automaton. Ion-beam induced mixing, which is governed by comparatively long-ranged interactions, leads to far-from-equilibrium processes (defect relaxation) producing disordered structures. A plain implementation of long-range interactions reduces the performance on GPUs substantially. In this talk we present an efficient implementation of long-range ballistic displacements in GPU-KLMC.
  • Lecture (Conference)
    2nd International Symposium “Computer Simulations on GPU”, 27.-29.05.2013, Freudenstadt, Deutschland
  • Lecture (others)
    Wigner RCP Institute Seminar, 30.08.2013, MTA-TTK, Budapest, Hungary

Registration No. 19037 - Permalink


Publication fees for Open Access journals - work flows at Helmholtz Centres
Reschke, E.; Köhler, M.; Wagner, A.; Grosse, K.
Open Access is a new way of publishing. The presentation aims to show how the golden way of Open Access Publishing has to be funding. Based on the "Berlin Declaration to Knowledge in the Science and Humanities" and the "Agreement of the Assembly of Members" (assembly of the directors of the Helmholtz Centres) the Helmholtz Research Centres organized the publishing and funding process in different ways. The presentation shows 4 examples for it: DESY, Forschungszentrum Jülich, GSI Darmstadt and HZDR.
Keywords: Open Access, Funding, Author fees, Publication fees,
  • Lecture (others)
    Open Access to Publications and Data in the Research Field "Structure of Matter" of the Helmholtz Association, 10.-11.06.2013, Hamburg, Deutschland

Registration No. 18877 - Permalink


Extremely Large Scale Simulation of Surface Growth and Lattice Gases
Ódor, G.; Schulz, H.; Kelling, J.; Heinig, K.-H.; Máté Ferenc, N.
Learn how to do efficient simulations of interface growth and Kinetic Monte Carlo on GPUs. Due to a mapping of the surface dynamics onto a binary lattices gas bit-coding permits to achieve simulations of unprecedentedly large sizes (2^17 x 2^17) and large speedups (~430 with respect to a single CPU). Explore new techniques in different implementations on GPUs using CUDA and OpenCL. Dive deep into the run-time and scaling analysis on different architectures to find optimal solutions for solving current simulation problems in the field of statistical physics and materials science.
Keywords: Computational Physics; Manufacturing Technical
  • Lecture (Conference)
    GPU Technology Conference, 18.-21.03.2013, San Jose, USA

Registration No. 18424 - Permalink


Large Scale Atomistic Simulations on Nanostructure Evolution
Kelling, J.; Heinig, K.-H.
The Kinetic Metropolis Lattice Monte-Carlo (KMC) method is a means of performing atomistic simulations of self-organization processes in solids at by far larger scales than those accessible via Molecular Dynamics (MD). This method was implemented on modern GPUs, which currently provide the most peek processing performance regarding both cost and energy consumption, achieving up to 70 times higher performance than the sequential reference implementation on a single core of a modern CPU. This enables atomistic simulations at even larger scales, even putting space and time scales comparable to the experiment within range.
  • Poster
    Response Treatment for the Dynamical Properties of Materials with the ABINIT Package, 22.-26.10.2012, ETH Zürich, Schweiz
  • Poster
    DPG-Frühjahrstagung der Sektion Kondensierte Materie (SKM), 10.-15.03.2013, Regensburg, Germany

Registration No. 18024 - Permalink


GPGPU for Kinetic Lattice Monte Carlo Simulations
Kelling, J.
Kinetic lattice Monte Carlo (KLMC) methods have for long been applied successfully for the atomistic large scale simulations of processes like annealing or ion beam treatment as well as for the study of scaling phenomena. Increasing the size of simulated systems is indispensable to get closer to the experiment as well as obtaining more accurate numerical estimates. In a time of stagnating sequential performance in computers parallelization is the only way to reach this goal. We have successfully taken on the implementation of two different methods for GPUs. In this talk I will outline the potential of general purpose processing on GPUs (GPGPU) for KLMC methods alongside presenting some of our key results.
  • Lecture (others)
    CCoE Seminar, 11.10.2012, TU-Dresden WIL/A317, Deutschland

Registration No. 17875 - Permalink


Library services of the Helmholtz-Zentrum Dresden-Rossendorf
Reschke, E.
In this slide show some topics of the library services will be presented, for example search for books, for journal articles, to use the SFX Linking service, how to use the licensed datatbasaes from home, publishing services of the library.
Keywords: Online catalogue, interlibrary loan, Linking services, document request, copyright, open access
  • Invited lecture (Conferences)
    7. Doktorandenseminar im IFA Schöneck, 08.-10.10.2012, Schöneck/Vogtl., Germany

Registration No. 17858 - Permalink


Utilisation of a Hybrid Approach for Immersive Industrial Process Control Visualisation
Skripcak, T.; Tanuska, P.; Schmeißer, N.
The objective of this paper is to present a proposal of hybrid system architecture for industrial process control visualisation. Leveraging of interoperability standard protocols for industrial automation, it is possible to create process agnostic visualisations. An application of this approach will lead to a hybrid system, which provides HMI (Human Machine Interface) layer for simulation, as well as real-world based industrial processes. An immersive 3-D visualisation is examined for the purpose of complex real-time process variables exploration. Reasonable 3-D visualisation types for industrial process HMI are identified. The idea is to create an immersive virtual environment for displaying industrial process related information, in order to help human operators to understand observed system behaviour and support them in the decision making process.
Keywords: SCADA/HMI, n-D visualisation, OPC, simulation.
  • Contribution to proceedings
    International Conference on Internet and Multimedia Technologies (ICIMT'12), 24.-26.10.2012, San Francisco, USA
    Proceedings of International Conference on Internet and Multimedia Technologies 2012., Hong Kong: International Association of Engineers, 978-988-19251-6-9, 415-420

Registration No. 17743 - Permalink


Cloud Computing and Big Data in German Research Institutes
Konrad, U.
Supercomputing and Big Data expand the spectrum of scientific studies, since they provide access to the understanding of phenomena of highest complexity in the area of the theories and models as well as for the analysis of experimental data. Since 2008 Cloud Computing has become another buzzword after Web 2.0. In the commercial area Cloud Computing hast already achieved an enormous significance. This plenary speech describes the fundamentals and the actual status of Big Data and Cloud Computing in the area of research in Europe.
Keywords: Information Technology, Supercomputing, Cloud Computing, Big Data
  • Invited lecture (Conferences)
    The 1st International Workshop for Young Scientists “Intelligent Technologies for Information Processing and Management”, 17.-19.07.2012, Ufa, Russia

Registration No. 17669 - Permalink


Accelerated Atomistic Simulations of Nanostructure Evolution using Graphics Cards
Kelling, J.
  • Lecture (others)
    Seminar Topical Problems, 16.05.2012, Chemnitz, Deutschland

Registration No. 17357 - Permalink


The Factor Cloud Computing in IT Risk Management
Penzina, V. J.; Konrad, U.; Yusupova, N. I.
Der Artikel beschreibt eine Definition und den Prozess des IT Risiko-Managements mit dem besonderen Einfluss von Cloud-Computing, insbesondere die Identifikation von Schwachstellen, Steuermöglichkeiten und Maßnahmen.
Keywords: Today Risk Management plays a crucial role in protecting information. Effective Risk Management is one of the most important parts of a security program in IT organizations. This paper represents a definition of Risk Management, first steps of the process and the influence of cloud computing according to vulnerability identification and controls and measures analysis subprocesses.
  • Contribution to proceedings
    The 14th International Workshop on Computer Science and Information Technologies, CSIT’2012, 20.-26.09.2012, MSC Lirica, Norway
    Proceedings of the 14th International workshop on computer science and information technologies, CSIT’2012, Ufa – Hamburg – Norwegian Fjords, 2012

Registration No. 17199 - Permalink


Open Access - Publizieren im Helmholtz-Zentrum Dresden-Rossendorf
Reschke, E.
Im Rahmen des Frühjahrstreffen der EU-Referenten der Helmholtz-Gemeinschaft im Helmholtz-Zentrum Dresden-Rossendorf am 08.u. 09.Mai 2012 werden die HZDR-Aktivitäten zu OPen Access dargestellt.
Keywords: Open Access, Open Data, Helmholtz-Zentrum Dresden Rossendorf, HZDR, Bibliothek, HORIZON 2020, EU Open Access Pilot; Special Clause 39,
  • Invited lecture (Conferences)
    Frühjahrstreffen EU Referenten der Helmholtz-Gemeinschaft, 08.05.2012, Dresden, Deutschland

Registration No. 17086 - Permalink


Comparison of Different Parallel Implementaions of the 2+1-Dimensional KPZ Model and the 3-Dimensional KMC Model
Kelling, J.; Ódor, G.; Nagy, M. F.; Schulz, H.; Heinig, K.-H.
We show that efficient simulations of the Kardar-Parisi-Zhang interface growth in 2 + 1 dimensions and of the 3-dimensional Kinetic Monte Carlo of thermally activated diffusion can be realized both on GPUs and modern CPUs. In this article we present results of different implementations on GPUs using CUDA and OpenCL and also on CPUs using OpenCL and MPI. We investigate the runtime and scaling behavior on different architectures to find optimal solutions for solving current simulation problems in the field of statistical physics and materials science.

Registration No. 17032 - Permalink


Kinetic Monte Carlo Simulations on Self-organization of Nanostructures Accelerated by Massive Parallelization
Kelling, J.
Modern graphics processing units (GPUs) currently provide the most peek processing performance regarding both cost and energy consumption. This is achieved by mounting large numbers of simple cores rather than a few complex ones. The characteristics that come with this design demand a large degree of data-parallelism from applications. Thus, new approaches are needed for parallelizing tasks that are not by nature data-parallel.
The 3D kinetic lattice Monte Carlo (KLMC) method is a means of performing atomistic simulations of self-organization processes in solids at by far larger scales than those accessible via molecular dynamics (MD). This method has been implemented for GPUs, achieving up to 70 times higher performance than the sequential reference implementation on a single core of a modern CPU. This enables atomistic simulations at even larger scales, even putting space and time scales comparable to the experiment within range.
The new program has been shown to be useful to study the phase separation in large binary systems. This was done with an application for third generation photovoltaics cells in mind which is a subject of a current BMBF project. It was also applied to out-of-equilibrium problems, backing up a theory of inverse Ostwald ripening (IOR) from an angle that was not previously looked at.
  • Diploma thesis
    TU Dresden, 2012
    62 Seiten

Registration No. 17024 - Permalink


Extremely large-scale simulation of a Kardar-Parisi-Zhang model using graphics cards
Kelling, J.; Ódor, G.
The octahedron model introduced recently has been implemented onto graphics cards, which permits extremely large-scale simulations via binary lattice gases and bit-coded algorithms. We confirm scaling behavior belonging to the two-dimensional Kardar-Parisi-Zhang universality class and find a surface growth exponent: β=0.2415(15) on 217×217 systems, ruling out β=1/4 suggested by field theory. The maximum speedup with respect to a single CPU is 240. The steady state has been analyzed by finite-size scaling and a growth exponent α=0.393(4) is found. Correction-to-scaling-exponent are computed and the power-spectrum density of the steady state is determined. We calculate the universal scaling functions and cumulants and show that the limit distribution can be obtained by the sizes considered. We provide numerical fitting for the small and large tail behavior of the steady-state scaling function of the interface width.
Keywords: 05.70.Ln, 05.70.Np, 82.20.Wt

Registration No. 16482 - Permalink


Innovative Informationstechnik im Dienste der Forschung am Beispiel des Helmholtz-Zentrums Dresden-Rossendorf.
Konrad, U.
Das Papier beschreibt innovative Anwendungen der Informationstechnologie in der Forschung am Helmholtz-Zentrum Dresden-Rossendorf (HZDR). Dazu werden Beispiele aus den Gebieten High Performance Computing, Visualisierung und Virtual Reality sowie Identity Management erläutert, bei denen das HZDR zusammen mit Partnern fortschrittliche Lösungen entwickelt hat.
Keywords: Information Technology, High Performance Computing, Visualization, Virtual Reality, Identity Management, Research Infrastructure
  • Invited lecture (Conferences)
    Regenerative Energien als Zukunftstechnologien, Internationale Tagung anlässlich des 120-jährigen Jubiläums der Ingenieurausbildung am Standort Köthen, 13.-14.10.2011, Köthen, Deutschland

Registration No. 16429 - Permalink


Interactive calibration and registration of electromagnetic trackung systems for Virtual Reality
Skripcak, T.; Tanuska, P.; Schmeißer, N.
This article is aimed on the calibration and registration of electromagnetic tracking system for the purpose of immersive virtual reality application. The first part of the article provides a basic introduction into position and orientation tracking subsystems and denotes problems which are occurring while electromagnetic approach is used for tracking purposes. After that two methods for tracking system calibration are presented and also a solution for position registration of tracking systems with projection space is provided. Second part is focused on interactive calibration application, where the process of development and a resulting application structure is described. Finally results obtained by usage of interactive calibration application are compared. The main goal of this article is to present a way of dealing with static errors in virtual reality application in order to obtain proper alignment of the virtual and real world.
Keywords: virtual reality, electromagnetic tracking system, calibration, registration
  • Contribution to proceedings
    IDS - INTERNATIONAL DOCTORAL SEMINAR, 15.-17.5.2011, Smolenice, Slovakia
    International Doctoral Seminar 2011 : Proceeding. Smolenice Castle, SR, Trnava: AlumniPress, 978-80-8096-145-9, 365-375

Registration No. 16273 - Permalink


Design and Implementation of Interactive Visualisation Configuration using Interaction Paradigms in Virtual Reality Environment
Skripcak, T.; Tanuska, P.; Schmeißer, N.
This article is aimed on the specific task of interaction with an immerse visualisation application. The first part of the article provides basic introduction into interaction paradigms in 3D space. After that classification and description of standard interaction tasks are presented. We introduce our view of matter on relations between the 3D interaction and standard interaction techniques. The second part describes hardware and software components of our VR system. Furthermore an overview of the architecture and implementation details of system for interactive visualisation configuration is discussed. We describe design specifications of a 3D UI, which helps to make interaction less error prone for inexperienced users. A specific solution for performing numerical input is also provided. The main goal of the article is to describe how 3D user interface paradigms can be implemented in the VR system.
Keywords: VirtualReality visualization, 3d interaction
  • Open Access LogoInternational Journal of Soft Computing and Engineering (IJSCE) 1(2011)5, 57-65

Registration No. 16272 - Permalink


An Approach for IT Service Management of Research Institutes
Konrad, U.; Lemke, C.
In times of scarce resources and increasing demands, an optimum IT support requires a greater service orientation and a more systematic description and management. While IT Service Management (ITSM) previously in the industry has become a standard, there are so far, only few best practice examples at universities and research institutions. The essential characteristic of this paradigm shift in the way of IT provisioning is to align IT with the specific business processes of the organization.

This paper proposes an approach to ITSM for research institutes that achieves an IT and business alignment in the context of an IT service orientation, based on ITIL v3. Starting with a description of the business processes in the context of research institutions the required IT services structure can be derived. A draft of a generic business process map providing the basis for the definition and design of IT services has been found. The resulting IT service and IT function catalogues combine the business and technical view. This provides the first step towards the development of a standardized and efficient implementation reference model for ITSM in the research area. The goal of this ongoing project is to develop a practical framework for the ITSM of research institutes and an IT management information system prototype.
Keywords: IT Service Management, ITIL v3, IT service orientation, business alignment
  • Contribution to proceedings
    Innovation Information Technologies: Theory and Practices, 08.-13.04.2011, Ufa, Russia
    Proceedings of the International Workshop "Innovation Information Technologies: Theory and Practice", 978-5-4221-0159-7, 1-4

Registration No. 15366 - Permalink


Simulation of 1+1 dimensional surface growth and lattices gases using GPUs
Schulz, H.; Ódor, G.; Ódor, G.; Nagy, M. F.
Restricted solid on solid surface growth models can be mapped onto binary lattice gases. We show that efficient simulation algorithms can be realized on GPUs either by CUDA or by OpenCL programming. We consider a deposition/evaporation model following Kardar-Parisi-Zhang growth in 1+1 dimensions related to the Asymmetric Simple Exclusion Process and show that for sizes, that fit into the shared memory of GPUs one can achieve the maximum parallelization speedup (~ ×100 for a Quadro FX 5800 graphics card with respect to a single CPU of 2.67 GHz). This permits us to study the effect of quenched columnar disorder, requiring extremely long simulation times. We compare the CUDA realization with an OpenCL implementation designed for processor clusters via MPI. A two-lane traffic model with randomized turning points is also realized and the dynamical behavior has been investigated.
Keywords: surface growth model, lattice gas, GPU, parallel algorithm

Registration No. 14949 - Permalink


Innovation Information Technology in the service of research at the FZD
Konrad, U.
Large research facilities, such as the Forschungszentrum Dresden-Rossendorf (FZD), provide the scientific work environment for employees, guests and students from all over the world.
To manage research projects, achieve new insights and visualize the results innovative information technology has to be applied for the communication, information and collaboration infrastructure as well as for the scientific equipments. Managing such a dynamic environment in a networked world is a challenging task. This paper describes some solutions and implementation aspects of innovative information technology for a large scale research facility developed in Dresden. The solutions are illustrated using some examples out of the areas of High Performance Computing (HPC), Virtual Reality (VR) and Collaboration in Virtual Institutes.
  • Contribution to proceedings
    International Workshop "Innovation Information Technologies: Theory and Practice", 06.-10.09.2010, Dresden, Deutschland
    Proceedings of the International Workshop "Innovation Information Technologies: Theory and Practice", Dresden: Forschungszentrum Dresden-Rossendorf, 978-3-941405-10-3, 10-12

Registration No. 14494 - Permalink


A Virtual Firefighting Simulator
Schmeißer, N.
A simulator will be presented which employs VR techniques in order to provide a training environment for re-ghters. While existing systems for re-brigade training do focus on tactics and command training, this simulator targets the training of the re attack group. It provides scenarios for search & rescue operations, direct reghting using water and foam as well as radiation source search operations. Visual, audio and haptic feed-back is provided. The system supports immersive displays (HMD) as well as standard techniques (monoscopic big-wall). It implements a walking paradigm, which enables the trainee to move around freely. The simulator uses extended reality features in order to improve the handling (nozzles, dose meter, ...). The trainee can use the system fully equipped, even using the respiratory protection, in order to provide a realistic feeling. Several physical models are used to provide eects closed to reality, namely:
  • heat release rate is used to model the energetic behavior of a fire, fl ame size etc.
  • a simple energetic model for re-extinction by cooling
  • water and foam ejection (depending on pressure, volume, form of jet, agent)
  • a smoke system (visibility equivalent to height of eye)
  • superposition of radiation sources (no shielding) and detection by Teleprobe FH40G
The simulation is supervised by an instructor, each action, e.g. radio traffc, movements is logged for analysis.
Keywords: simulation fire fire-fighting extinction radioton search rescue
  • Contribution to proceedings
    3rd International Workshop "Innovation in Information Technologies - Theory and Practice", September 6th-10th, Dresden, Germany, 2010, 06.-10.9.2010, Dresden, BRD
    A Virtual Firefighting Simulator, Dresden: Forschungszentrum-Rossendorf e. V., 978-3-941405-10-3, 76-82

Registration No. 14468 - Permalink


Computing the KPZ Equation Using GPU Acceleration
Schulz, H.; Odor, G.; Kelling, J.; Heinig, K.-H.; Liedke, B.; Schmeißer, N.
In this talk we present an approach for fast parallel computation of the Kardar-Parisi-Zhang equation (KPZ). Acceleration of these computations is always a major task since the run time of the simulations representing sequential algorithms is in the range of several months. We developed an implementation for the KPZ equation on graphics processing units (GPU) using the Compute Unified Device Architecture (CUDA). Experimental results show a significant speed-up compared to implementations on conventional processors.
  • Contribution to proceedings
    Innovation Information Technologies - Theory and Practice, 06.-10.09.2010, Dresden, Deutschland
    Computing the KPZ Equation Using GPU Acceleration, Dresden: FZD, 978-3-941405-10-3, 46-48

Registration No. 14463 - Permalink


Identity Management in Large Research Facilities
Konrad, U.; Schmeißer, N.
Large research facilities, such as the Forschungszentrum Dresden-Rossendorf, provide a scientific work environment for employees, guests and students from all over the world. All these users require access to local and remote facilities, to the scientific equipment and the communication, information and collaboration infrastructure. Nearly all projects currently running are using state of the art web and GRID technologies crossing borders and continents with people joining in and leaving. Managing a dynamic environment like this is a challenging task. This paper describes the concepts and implementation aspects of an Identity Management System (IDM) for a large scale research facility.
  • Contribution to proceedings
    The Russian-German Workshop "Innovation Information technologies: theory and practice", 26.-31.07.2009, Ufa, Russia
    Identity Management in Large Research Facilities, UFA: Eigenverlag UFA State Technical University

Registration No. 12980 - Permalink


Polyhedral Surface Approximation of Non-Convex Voxel Sets and Improvements to the Convex Hull Computing Method
Schulz, H.
In this paper we introduce an algorithm for the creation of polyhedral approximations for objects represented as strongly connected sets of voxels in three-dimensional binary images. The algorithm generates the convex hull of a given object and modifies the hull afterwards by recursive repetitions of generating convex hulls of subsets of the given voxel set or subsets of the background voxels. The result of this method is a polyhedron which separates object voxels from background voxels. The objects processed by this algorithm and also the background voxel components inside the convex hull of the objects are restricted to have genus 0. The second aim of this paper is to present some improvements to our convex hull algorithm to reduce computation time.
Keywords: digital geometry, surface approximation, abstract cell complex, polyhedron, voxel
  • Open Access LogoWissenschaftlich-Technische Berichte / Forschungszentrum Dresden-Rossendorf; FZD-514 2009

Downloads:

Registration No. 12497 - Permalink


Library Information Services - not only books in shelves
Reschke, E.
The presentation shows the Library information services, focused on the elctronic services within the intranet and the internet. The main point was the link resolver for linking services.
Keywords: online catalogue, document deliver, Interlibrary loan, elctronic databases, virtual library, publishing, copyright, document managment systems, open access
  • Lecture (others)
    FZD Lecture, 01.12.2008, Dresden, Deutschland

Registration No. 11997 - Permalink


Polyhedral Approximation and Practical Convex Hull Algorithm for Certain Classes of Voxel Sets
Schulz, H.
In this paper we introduce an algorithm for the creation of polyhedral approximations for certain kinds of digital objects in a three-dimensional space. The objects are sets of voxels represented as strongly connected subsets of an abstract cell complex. The proposed algorithm generates the convex hull of a given object and modifies the hull afterwards by recursive repetitions of generating convex hulls of subsets of the given voxel set or subsets of the background voxels. The result of this method is a polyhedron which separates object voxels from background voxels. The objects processed by this algorithm and also the background voxel components inside the convex hull of the objects are restricted to have genus 0. The second aim of this paper is to present some practical improvements to the discussed convex hull algorithm to reduce computation time.
Keywords: digital geometry, convex hull, abstract cell complex, abstract polyhedron, surface approximation

Registration No. 11658 - Permalink


Multi-Vendor Enterprise Storage-Lösungen erfolgreich integrieren
Konrad, U.
Eingeladener Vortrag der Computerwoche Fachkonferenz: Data Center up-to-date 08 - virtualisieren, konsolidieren, outsourcen
Es wird ein Anwendervortrag gehalten, der Fokus liegt auf folgenden Schwerpunkten:
- IT-Profil des Forschungszentrums Dresden-Rossendorf
- Enterprise Storagelösungen
- Virtualisierung im Speichernetzwerk
- Einsatz von Filesystemen
- Lebenszyklus eines Storagesystems
Ein wesentlicher Punkt des Vortrages ist die Beschreibung der am FZD geschaffenen Lösung mit in-band Virtualisierung des Speichers im Netzwerk über spezielle Appliances sowie die Auswertung der Erfahrungen dazu.
Keywords: information technology, data center, enterprise storage solutions, in-band virtualization, SAN, file systems, computerwoche
  • Invited lecture (Conferences)
    Data Center up-to-date 08 - virtualisieren, konsolidieren, outsourcen, 10.04.2008, Frankfurt am Main, Germany

Registration No. 11203 - Permalink


Das neue Urheberrecht seit 1.1.2008, Dokumentenbereitstellung, SUBITO, EZB - Erfahrungen aus der Praxis
Reschke, E.
Die Umsetzung des neuen Urheberechts für den Bereich der Dokumentenbereitstellung durch Bibliotheken ist äußerst problematisch. Die Einschränkungen der Verfügbarkeit elektronischer Lieferungen bedingen umfangreiche Schulungen des Bibliothekspersonals, Informationen der Bibliotheksbenutzer, zusätzlichen Bearbeitungsaufwand der Bestellungen und Bestelleingangs- und Rechnungsverwaltung und wesentlichen finanziellen Mehraufwand. Im Vortrag wird am Beispiel des Subito-Rahmenvertrages auf die aktuellen Probleme eingegangen.
  • Lecture (others)
    Treffen des Arbeitskreises Spezialbibliotheken Dresden, 02.04.2008, Dresden, Germany

Registration No. 11192 - Permalink


Polyhedral Surface Approximation of Non-Convex Voxel Sets through the Modification of Convex Hulls
Schulz, H.
In this paper we want to introduce an algorithm for the creation of polyhedral approximations for objects represented as strongly connected sets of voxels in three-dimensional binary images. The algorithm generates the convex hull of a given object and modifies the hull afterwards by recursive repetitions of generating convex hulls of subsets of the given voxel set or subsets of the background voxels. The result of this method is a polyhedron which separates object voxels from background voxels. The objects processed by this algorithm and also the background voxel components inside the convex hull of the objects are restricted to have genus 0.
Keywords: Digital Geometry, Convex Hull, Surface Approximation, Abstract Polyhedron
  • Contribution to proceedings
    12th International Workshop on Combinatorial Image Analysis, 07.-09.04.2008, Buffalo, NY, USA
    Combinatorial Image Analysis

Registration No. 10819 - Permalink


Requirements on the HLD-Grid
Kunicke, M.; Herrmannsdörfer, T.
In 2007 the Dresden High Magnetic Field Laboratory (HLD) [1] will come into operation. This collective project of five institutes located in Dresden will used by the high field community in Europe. The IT tasks concerning the shared use of the experimental facilities should be solved using Grid technologies. The report describes the requirements on the HLD-Grid.
Keywords: Grid, shared use, experimental facility, High Magnetic Field Laboratory
  • Lecture (Conference)
    Distributed computing and Grid technologies in science and education GRID'2006, 26.-30.06.2006, Dubna, Russia
  • Contribution to proceedings
    Distributed Computing and Grid-Technologies in Science and Education GRID'2006, 26.-30.06.2006, Dubna, Russia
    Proceedings of the Second International Conference Distributed Computing and Grid-Technologies in Science and Education, 5-9530-0138-X, 113-116

Registration No. 8508 - Permalink


Efficient Determination of Optimal Regularization Parameter for Inverse Problem in EXAFS Spectroscopy
Kunicke, M.; Kaminsky, I. Y.; Babanov, Y. A.; Funke, H.
This paper presents a method for choosing the optimal regularization parameter in EXAFS analysis based on the L-curve-criterion without any external pa-rameter. The L-curve is a log-log plot of the norm of regularized solution ver-sus the norm of the corresponding residual norm. The optimal regularization parameter is chosen as a maximum of the L curve curvature. We apply this method in two experimental cases: crystalline Cu with well-known structure and aqua-ion Cm+3 with unknown structure. The optimal regularization pa-rameter leads to a stable solution with a good convergence of the iterative process
  • Physica Scripta T115(2005), 237-239

Downloads:

Registration No. 8011 - Permalink


Neues aus der Bibliothek
Reschke, E.
Die Bibliothek ist die zentrale Stelle für wissenschaftliche Literatur- und Fachinformation des Forschungszentrums Rossendorf. Mit ihren spezifischen Leistungen unterstützt sie die Forschung und trägt damit zur Sicherung der Attraktivität des Forschungsstandortes Rossendorf im internationalen Standortwettbewerb bei. Davon ausgehend werden die neuesten elektronischen Informationsservices dargestellt: Zugang zum "Web of Knowledge" und zu einer Vielzahl elektronischer Zeitschriften im Volltext. Durch Umbau der ehemaligen Buchbinderei im Haus der Bibliothek konnte ein IT-Schulungsraum eingerichtet werden.
Keywords: Library, Information services, Database availability,
  • FZR intern (2004) Februar

Registration No. 6072 - Permalink