industrial visualisation
Windows, Mac, Unix and Linux operating systems, and it does shared-memory parallel processing on up to 64 cores.
Open source alternatives Besides these commercial programs, Boston University’s Brisson points out several interesting open source alternatives. First is the Visualisation Toolkit (VTK), which is a set of graphics libraries accessible from C++, Tcl, Perl, Python or Java. The developers’ intent is that VTK code be platform independent. Its graphics interaction windows provide flexibility, and there are a number of interaction/navigation options. Built on top of VTK, ParaView is an
A plot created with Origin shows a Gaussian surface fit to 3D scatter points, transparency of the surface and ‘in-plane’ axis tick labels
datasets. In addition, full NAG capability is built in. For connectivity, there is also a COM interface to popular Office programs. Further, with one mouse click you can send all the graphs in a project to a single PowerPoint file. As for 3D plots, Origin does all traditional
xyz plots, and for 3D surfaces it works with a z-array. When it comes to the size of projects, all data must fit in local memory at one time, but work is ongoing to handle large data files without loading everything into RAM. ‘The amount of data today is staggering,’ says Laurent Bernardin, executive VP of R&D and chief scientist at Maplesoft, ‘and visualisation is the key to making sense of all this data. In the past, visualisation was a batch process where you would see your pictures the next day. Today, people want instant gratification, they want to “play” with the data and explore it interactively. This all comes down to the user interface and processing power. Thus, the big trends we see are parallelisation, multicore support and GPUs.’ The Maple package has a front end
which focuses on interactive elements for visualising and exploring data. It has all the same numerical abilities as packages such as Matlab, but besides using commands, users can build symbolic equation-based models as well as interactive live documents with elements such as buttons, sliders, and live graph regions.
44 SCIENTIFIC COMPUTING WORLD Also interesting are interactive links
to CAD packages, so that after pulling parameters into Maple, users perform an analysis, find their optimal values and press a button to have that new value appear in the CAD drawing. Another recent feature is network connectivity with the Grid Computing Toolbox, whereby you can import data with TCP/IP or other data flows. And while all data being evaluated must fit into local RAM, you can distribute a large job among many nodes in a shared-memory system and run tasks in parallel. For GPUs, Maple supports Cuda, but this is not suited for all tasks because of the bottleneck of moving data from local RAM into GPU memory. While many popular plotting/
visualisation packages don’t handle images from detailed simulations such as computational fluid dynamics, Tecplot has augmented its standard Tecplot Focus product with Tecplot 360. It supports 32 different CFD, FEA, structural analysis and general-purpose data formats including CGNS, Flow-3D, Fluent, Plot3D, and Star-CCM. A CFD analyser examines grid quality, performs spatial integration, generates particle trajectories, extracts flow features and estimates numerical errors. For exploration, it has a slicing tool, iso-surface tool, streamtrace tour and contour tool. The software can animate and step through transient solutions. The software runs on
open source visualisation application whose greatest strength is as an interactive visualisation tool, and Brisson points out that it’s easy to quickly develop such a complex and effective visualisation. The array of tools provided (via VTK) and the ease of interaction make this the tool a good option for non-programming visualisation. Data exploration can be done interactively in 3D or programmatically using ParaView’s batch processing capabilities. Important for HPC users, ParaView was developed to analyse extremely large datasets using distributed memory computing resources. It can be run on supercomputers to analyse datasets of terascale as well as on laptops for smaller data. Those who want professional support
for ParaView can turn to Kitware. Among that company’s other products is ActiViz, which generates C# wrappers around VTK, enables developers to combine the power of VTK with the many .NET framework objects for web and database access. VolView is an interactive system for volume visualisation; 3D tools include volume rendering, maximum intensity projections, and oblique reformatting. Of special interest to HPC users is Midas, which integrates multimedia server technology with Kitware’s open- source analysis and visualisation clients. It has been optimised for storing massive collections of scientific data and related metadata and reports.
References 1. ‘SCV Tutorials’, Scientific Computing and Visualization Group, Boston University,
www.bu.edu/tech/research/training/ tutorials/list/ 2. ‘Tutorial: Color in Scientific Visualization’, Department of Computer Science College of Engineering and Applied Science, University of Colorado Boulder, https://
csel.cs.colorado.edu/~csci4576/SciVis/
SciVisColor.html
www.scientific-computing.com
Page 1 |
Page 2 |
Page 3 |
Page 4 |
Page 5 |
Page 6 |
Page 7 |
Page 8 |
Page 9 |
Page 10 |
Page 11 |
Page 12 |
Page 13 |
Page 14 |
Page 15 |
Page 16 |
Page 17 |
Page 18 |
Page 19 |
Page 20 |
Page 21 |
Page 22 |
Page 23 |
Page 24 |
Page 25 |
Page 26 |
Page 27 |
Page 28 |
Page 29 |
Page 30 |
Page 31 |
Page 32 |
Page 33 |
Page 34 |
Page 35 |
Page 36 |
Page 37 |
Page 38 |
Page 39 |
Page 40 |
Page 41 |
Page 42 |
Page 43 |
Page 44 |
Page 45 |
Page 46 |
Page 47 |
Page 48 |
Page 49 |
Page 50 |
Page 51 |
Page 52