Visualization. Discovering immersive and interactive 3D simulation environments for data analysis and integrated software frameworks.

Transcription

1 Visualization Discovering immersive and interactive 3D simulation environments for data analysis and integrated software frameworks.

2 2

3 HLRS High Performance Computing Center Stuttgart The High Performance Computing Center of Stuttgart (HLRS) of the University of Stuttgart is the first National Supercomputing Center in Germany and is offering services to both academic users and industry. Apart from the operation of supercomputers HLRS activities include teaching and training in distributed systems, software engineering and programming models, as well as development of new technologies. HLRS is an active player in the European research arena with special focus on Scientific Excellence and Industrial Leadership initiatives. Our Network: HLRS is tightly connected to academia and industry through long term partnerships with global market players such as Porsche and T-Systems, as well as worldwide companies, HPC centres and Universities. Particular attention is given to collaboration with Small and Medium Enterprises (SMEs). Our Experience: HLRS has been at the forefront of regional, national and European research and innovation over the last 20 years. During this time, HLRS has participated successfully in more than 90 European research and innovation projects. Our Expertise: HLRS is a leading innovation center, applying software engineering methods to HPC and Cloud for the benefit of multiple application domains such as automotive, engineering, health, mobility, security, and energy. Thanks to the close interaction with industry, the center s capabilities and expertise support the whole lifecycle of simulation covering research aspects, pre-competitive development and preparation for production. The HLRS innovation group, which actively examines and tests new technologies, can bring into projects expertise on leading edge technologies hardware and scale up data analysis techniques. Our Infrastructure: HLRS operates a CRAY XC40 supercomputer (peak performance > 7 PetaFlops), as well as a variety of smaller systems, reaching from clusters to cloud resources. Director HLRS Prof. Dr. Michael Resch Programming Models & Tools Big Data, Analytics & Management Cloud Computing Optimization & Scalability Featured Topics Visualization Exascale Computing Energy Efficiency Services HLRS 3

4 Visualization Visualization has always been an important research tool for science and industry, as it enables researchers to depict results, problems, and optimization potential in a vivid, illuminating way which could not be provided by a pure sequence of bits and bytes. With the increasing complexity of simulation models and, in turn, the size of datasets, the simulation, analysis, and visualization process has become much more time-consuming. For complex engineering and science procedures that need high accuracy and reliability of their respective visualized outcomes, high-performance computing (HPC) serves as an indispensable tool by being able to process large and complex datasets in short time. To this end, the High-Performance Computing Center Stuttgart (HLRS) built the CAVE a cube with a side length of 2.7m that uses a large variety of virtual and augmented reality equipment. This enables researchers, developers but also society and politiciansto interact with a fully immersive, 3D simulation environment to analyze and discuss their computations or computations made for them. The floor of the CAVE is made of a three-layer laminated safety glass, covered by a rear projection screen, while the walls and ceiling are built of dark plexiglass screens. To make the best possible use of the in-house visualization infrastructure, HLRS researchers continually develop specific software environments, frameworks, and graphics toolkits that meet the requirements of an HPC system environment. Especially the COVISE toolkit and its successor VISTLE are taken up world-wide. With this brochure, we invite you to learn more about applied visualization in cases such as city planning and urban development. Project Overview Covise - A Collaborative and Distributed Software Environment for High Performance Simulation and Visualization Page 5 ODDLOT - Road Networks for Driving Simulators ODDLOT_ The OpenDrive Designer Page 6 Urban Lab City Quarters Reallabor Stadtquartiere 4.0 Page 7 VIRVO - Virtual Rality Volume Rendering Seite 8 Visdral - Virtual Investigation, Simulation and Documentation of Road Accidents via Laserscanning Seite 9 Vistle - A Software Environment for High-Performance Simulation and Parallel Visualization Seite 10 Visualization 4

5 Covise A Collaborative and Distributed Software Environment for High-Performance Simulation and Visualization COVISE, the COllaborative VIsualization and Simulation Environment is an extensible distributed software environment. It integrates simulations on supercomputers, post-processing and collaborative visualization functionalities in a seamless manner. COVISE applications are composed of multiple tasks which are modeled as operating system processes. The corresponding work flows are configured in a graphical pipeline editor and can span different machines. COVISE allows several users to work in a collaborative synchronized session, both on desktop systems and with OpenCOVER in immersive virtual environments. The development of COVISE was initiated in European Community funded projects in cooperation with industry. COVISE is licensed under the GNU Lesser General Public License and available on GitHub. Its developmentis led by HLRS in a continuous effort since Supported Platforms: Windows Linux Mac OS COVISE has shown its versatility with many commercial users and in research projects at the High Performance Computing Center Stuttgart (HLRS) and its partners. It fulfills the needs of diverse branches such as automotive industry, architecture, hydraulic machinery, and archeology. More than 50 file formats such as Ensight, CFX, and StarCD are supported. Contact Dr. Uwe Wössner Phone: +49 (0) 711/ woessner@hlrs.de Further Information projects/covise/ Covise 5

6 ODDLOT Road Networks for Driving Simulators ODDLOT: The OpenDRIVE Designer Current research in the field of autonomous driving, the development of advanced driver assistance systems (ADAS) and new regulations for the safety of autonomous features, require extensive testing in virtual environments and simulators. Roads, signals and road objects for driving simulators have to be modelled as close to reality as possible. A lot of time is necessary to collect data and model the road network, the landscape, vegetation and road signs and traffic lights. A standard file format has the advantage of easy exchange between different simulation environments, applications and companies. OpenDRIVE is a standard to describe a road network, including maps, objects, surface properties and signaling. Particular specifications for certain driving simulators can be added via user defined tags. In ODDLOT the user can load maps and draw roads in a 2-dimensional graphical interface using different prototypes, specified by the number of lanes, their width and road markings. Roads can be linked together, allowing to manipulate the driving decisions of autonomous intelligent vehicles in the scene. Junctions can be created automatically by defining the area of intersection or manually in a specialized editor. ODDLOT comes with a bundle of common German signals as well as Chinese signals to make the placement of signs easy. Different kinds of barriers are available too, but new objects can also be described in an XML file. Tiles can be connected to build large networks, thereby reducing the modelling time by reusing already existing tiles. To ease the creation of virtual driving simulator tracks the High Performance Computing Center at the University of Stuttgart (HLRS) is developing an OpenDRIVE Editor. The so called ODDLOT is part of COVISE, the Collaborative Visualization and Simulation Environment of the HLRS, an Open Source package (LGPL2+) available on GitHub. ODDLOT is adapting to the ongoing extensions of the OpenDRI- VE standard, making it possible to describe dynamically changing elements in the future. Contact Jutta Sauer/ Dr. Uwe Wössner Phone: +49 (0) 711/ (0) 711/ sauer@hlrs.de woessner@hlrs.de Further Information ODDLOT 6

7 Urban Lab City Quaters 4.0 Reallabor Stadtquartiere 4.0 The digitization of economy and society is one of the most comprehensive upheavals of our time. Ever-shorter innovation cycles of digital products and processes are conflicting with the longevity and path dependency of our cities and their infrastructures. Though, in many cases urban planning and decision-making processes are still carried out with conventional, highly regulated procedures and tools, that can hardly keep pace with current developments. The project follows a strategy of co-designing research and joint knowledge acquisition with citizens and other local actors. It is intended to develop forward-looking procedures and digital tools through a series of real-world experiments. These real-world experiments are conducted and evaluated in cooperation with the cities of Stuttgart and Herrenberg in ongoing urban planning projects. Urban Lab City Quarters 4.0 is funded by the Ministry of Science, Research and the Arts Baden-Württemberg. Project Partners: University of Stuttgart: Addressing the digitization of planning and decision-makingprocesses involving all relevant stakeholders, Urban Lab City Quarters 4.0 focuses questions such as: Stuttgart Research Center for Interdisciplinary Risk and Innovation Studies (ZIRIUS) Institute of Human Factors and Technolgy Management (IAT) Faculty of Architecture and Urban Design High Performance Computing Center Stuttgart (HLRS) Fraunhofer Institute for Industrial Engineering, Stuttgart Kommunikationsbüro Ulmer, Stuttgart The cities of Herrenberg and Stuttgart How can future-oriented influencing factors be reflected in urban planning models? What is the role of citizens and their local ex pertise in anticipating and planning the future? Which formats and tools, among them future scenarios, interactive and immersive visualizations as well as simulations, can help with the interface between ideas development and planning, citizenship and a city s authorities? Contact Dr. Uwe Wössner Phone: +49 (0) 711/ woessner@hlrs.de Further Information Urban Lab City Quarters 4.0 7

8 Virvo Virtual Reality Volume Rendering VIRVO is an open source framework for direct volume rendering. It is an integral part of the COVI- SE visualization and simulation environment, which was developed at the computing center of the University of Stuttgart. The development of VIRVO was initiated at HLRS within German Collaborative Research Centers 382 and 374. VIRVO is developed further within the DeskVox project on GitHub. VIRVO offers alternative techniques, which render the volume data on a remote and powerful parallel computer. The display machine receives only the rendered images and does not have to deal with a large amount of volume data. The shear-warp algorithm is used for rendering, which can render volume data very quickly and in software, without the need for cutting edge hardware on the display machine. For higher image fidelity, vectorized CPU ray casting can be used. Volume rendering is used to display three dimensional scalar datasets, for example the temperature distribution in a car cabin or proton densities in MRI scans (see image below).traditional approaches display only a subset of the dataset, most of the data can only be displayed by varying the selected parameters.with direct volume rendering, the entire dataset can be displayed in one image. Every scalar value is assigned both a color and a transparency by a user definable transfer function. VIRVO also has the capability to simultaneously display multiple channels of scalar values. The most frequently used technique to display volume data is by slicing the dataset and displaying it with a stack of textures (3D textures). This requires large texture memory on the display computer. Contact Dr. Uwe Wössner Phone: +49 (0) 711/ woessner@hlrs.de Further Information Virvo 8

9 Visdral Virtual Investigation, Simulation and Documentation of Road Accidents via Laserscanning Computer aided crash analysis is a central element for the identification of accident causes. It serves solving of legal issues as well as optimization of active and passive safety of vehicles. Furthermore, constructional measures at black spots for accidents can be taken to prevent accidents. 3D-scanners have become widely available for many applications. They allow capturing of location and vehicles but lead to massive amount of data in form of point clouds. Visdral aims at developing advanced methods in computer aided crash analysis by progressing data acquisition as well as data processing. These methods include virtually reconstructing the scene by describing road surface, environment, vehicles and persons involved in the accident in standardized formats. This paves the road to new simulation techniques building structural simulation models from 3D-scanned data. Figures 1 and 2: Virtual crash scene investigation combining 3D-scanned data with the results of a FEM simulation in a virtual interactive environment. Contact Dr. Uwe Wössner Phone: +49 (0) 711/ woessner@hlrs.de Further Information projects/visdral/ Visdral 9

10 Vistle A Software Environment for High-Performance Simulation and Parallel Visualization Vistle, the VISualization Testing Laboratory for Exascale computing, is an extensible software environment that integrates simulations on supercomputers, post-processing and parallel interactive visualization.it is under active development at HLRS since 2012 within the European project CRESTA and bwvisu. The objective is to provide a highly scalable successor to COVISE, exploiting data, task and pipeline parallelism in hybrid shared and distributed memory environments with accelerationhardware. Domain decompositions used during simulation can be reused for visualization.a Vistle work flow consists of several processing modules, each of which is a parallel MPI program that uses OpenMP within nodes. These can be configured graphically or from Python. available. Hybrid remote rendering allows to combine its output with local rendering, while ensuring smooth interactivity by decoupling it from remote rendering. The Vistle system is modular and can be extended easily with additional visualization algorithms. Source code is available on GitHub and licensed under the LPGL.The figure shows an interactive parallel visualization of a pump turbine. The simulation was conducted by IHS. Shared memory is used for transfering data between modules on a single node. Work flows can be distributed across several clusters. For rendering in immersive projection systems, Vistle uses Open- COVER. Visualization parameters can be manipulated from within the virtual environment. Large data sets can be displayed with OpenGL sort-last parallel rendering and depth compositing. For scaling with the simulation on remote HPC resources, a hybrid sort-last/sort first parallel ray casting renderer based on the Embree CPU raytracing framework is Contact Dr. Uwe Wössner Phone: +49 (0) 711/ woessner@hlrs.de Further Information Vistle 10

11

12 High Performance Computing Center Stuttgart (HLRS) Mail: University of Stuttgart Nobelstrasse Stuttgart Germany Phone: +49 (0)711 / Fax: +49 (0)711 / Editor: Lena Bühler, Eric Gedenk, Dr. Bastian Koller Design: Janine Jentsch, Ellen Ramminger HLRS 2017