Large Scale Visualization with ParaView

Supercomputing 2013 Tutorial

November 17, 2013

Kenneth Moreland W. Alan Scott
Sandia National Laboratories

David DeMarle
Kitware, Inc.

Li-Ta Lo
Los Alamos National Laboratory
Sandia National Laboratories is a multi-program laboratory managed and operated by Sandia
Corporation, a wholly owned subsidiary of Lockheed Martin Corporation, for the U.S. Department of
Energys National Nuclear Security Administration under contract DE-AC04-94AL85000.
SAND 2013-9368P
Outline
Introduction
Basic Usage
Visualizing Large Models
To Follow Along
Install ParaView 4.1.
http://www.paraview.org Download
Get example material.
http://www.paraview.org/Wiki/The_ParaView_Tutorial
Data also available on thumb drives.
Please dont steal the thumb drives.
They are personal property volunteered by presenters.
They have never touched a Sandia computer.
Introduction
What is ParaView?
An open-source, scalable, multi-
platform visualization application.
Support for distributed computation
models to process large data sets.
An open, flexible, and intuitive user
interface.
An extensible, modular architecture
based on open standards.
A flexible BSD 3 Clause license
Commercial maintenance and support.
Renato N. Elias, NACAD/COPPE/UFRJ , Rio de J anerio, Brazil
Swiss National Supercomputing Centre
J erry Clarke, US Army Research Laboratory
Current ParaView Usage
Used by academic, government, and
commercial institutions worldwide.
Downloaded ~100K times per year.
HPCwire Editors Choice 2010 and
HPCwire Readers Choice 2010/2012
Awards for Best Visualization Product or
Technology.

Data Ranges
Used for all ranges of data size.
Landmarks of SNL usage:
6 billion structured cells (2005).
250 million unstructured cells (2005).
Billions of AMR cells (2008).
Scaling test over 1 Trillion cells (2010).
ParaView Application Architecture
MPI OpenGL IceT Etc.
VTK
ParaView Server
ParaView Client pvpython Custom App
UI (Qt Widgets, Python Wrappings)
ParaWeb Catalyst
ParaView Development
Started in 2000 as collaborative effort
between Los Alamos National Laboratories
and Kitware Inc. (lead by J ames Ahrens).
ParaView 0.6 released October 2002.
September 2005: collaborative effort between
Sandia National Laboratories, Kitware Inc.
and CSimSoft to rewrite user interface to be
more user friendly and develop quantitative
analysis framework.
ParaView 3.0 released in May 2007.
ParaView 4.0 released in J une 2013.
Current Funding
ARL
ERDC
US Army (SBIR)
US Air Force (STTR)
ONR
Support Contracts
Electricity de France
Microsoft
Other contributors
Swiss National
Supercomputing Centre
DOE SLAC
Ohio State
Mississippi State
RPI
Basics of Visualization

Data Types
Uniform Rectilinear
(vtkImageData)
Non-Uniform Rectilinear
(vtkRectilinearData)
Curvilinear
(vtkStructuredData)
Polygonal
(vtkPolyData)
Unstructured Grid
(vtkUnstructuredGrid)
Multi-block
Hierarchical Adaptive
Mesh Refinement
(AMR)
Hierarchical Uniform
AMR
Octree
More Information
Online Help F1
The ParaView Users Guide
http://paraview.org/Wiki/ParaView/Users_Guide/Table_Of_C
ontents

The ParaView web page
www.paraview.org
ParaView mailing list
[email protected]
Basic Usage
User Interface
Menu Bar
Toolbars
Pipeline Browser
Properties Panel
3D View
Advanced Toggle
Getting Back GUI Components
Creating a Cylinder Source
1. Go to the Source menu and select
Cylinder.
2. Click the Apply button to accept the
default parameters.
Simple Camera Manipulation
Drag left, middle, right buttons for
rotate, pan, zoom.
Also use Shift, Ctrl, Alt modifiers.
Creating a Cylinder Source
1. Go to the Source menu and select
Cylinder.
2. Click the Apply button to accept the
default parameters.
3. Increase the Resolution parameter.

4. Click the button again.
Pipeline Object Controls
Undo Redo
Redo Undo
Camera
Redo
Camera
Undo
Render View Options
Display Properties
Creating a Cylinder Source
1. Go to the Source menu and select
Cylinder.
2. Click the Apply button to accept the
default parameters.
3. Increase the Resolution parameter.

4. Click the button again.
5. Delete the Cylinder.
Supported Data Types
ParaView Data (.pvd)
VTK (.vtp, .vtu, .vti, .vts, .vtr)
VTK Legacy (.vtk)
VTK Multi Block
(.vtm,.vtmb,.vtmg,.vthd,.vthb)
Partitioned VTK (.pvtu, .pvti,
.pvts, .pvtr)
ADAPT (.nc, .cdf, .elev, .ncd)
ANALYZE (.img, .hdr)
ANSYS (.inp)
AVS UCD (.inp)
BOV (.bov)
BYU (.g)
CAM NetCDF (.nc, .ncdf)
CCSM MTSD (.nc, .cdf, .elev,
.ncd)
CCSM STSD (.nc, .cdf, .elev,
.ncd)
CEAucd (.ucd, .inp)
CMAT (.cmat)
CML (.cml)
CTRL (.ctrl)
Chombo (.hdf5, .h5)
Claw (.claw)
Comma Separated Values
(.csv)
Cosmology Files (.cosmo,
.gadget2)
Curve2D (.curve, .ultra, .ult, .u)
DDCMD (.ddcmd)
Digital Elevation Map (.dem)
Dyna3D(.dyn)
EnSight (.case, .sos)
Enzo boundary and hierarchy
ExodusII (.g, .e, .exe, .ex2,
.ex2v.., etc)
ExtrudedVol (.exvol)
FVCOM (MTMD, MTSD,
Particle, STSD)
Facet Polygonal Data
Flash multiblock files
Fluent Case Files (.cas)
GGCM (.3df, .mer)
GTC (.h5)
GULP (.trg)
Gadget (.gadget)
Gaussian Cube File (.cube)
J PEG Image (.jpg, .jpeg)
LAMPPS Dump (.dump)
LAMPPS Structure Files
LODI (.nc, .cdf, .elev, .ncd)
LODI Particle (.nc, .cdf, .elev,
.ncd)
LS-DYNA (.k, .lsdyna, .d3plot,
d3plot)
M3DCl (.h5)
MFIX Unstructred Grid (.RES)
MM5 (.mm5)
MPAS NetCDF (.nc, .ncdf)
Meta Image (.mhd, .mha)
Miranda (.mir, .raw)
Multilevel 3d Plasma (.m3d,
.h5)
NASTRAN (.nas, .f06)
Nek5000 Files
Nrrd Raw Image (.nrrd, .nhdr)
OpenFOAM Files (.foam)
PATRAN (.neu)
PFLOTRAN (.h5)
PLOT2D (.p2d)
PLOT3D (.xyz, .q, .x, .vp3d)
PLY Polygonal File Format
PNG Image Files
POP Ocean Files
ParaDIS Files
Phasta Files (.pht)
Pixie Files (.h5)
ProSTAR (.cel, .vrt)
Protein Data Bank (.pdb, .ent,
.pdb)
Raw Image Files
Raw NRRD image files (.nrrd)
SAMRAI (.samrai)
SAR (.SAR, .sar)
SAS (.sasgeom, .sas,
.sasdata)
SESAME Tables

SLAC netCDF mesh and mode
data
SLAC netCDF particle data
Silo (.silo, .pdb)
Spheral (.spheral, .sv)
SpyPlot CTH
SpyPlot (.case)
SpyPlot History (.hscth)
Stereo Lithography (.stl)
TFT Files
TIFF Image Files
TSurf Files
Tecplot ASCII (.tec, .tp)
Tecplot Binary (.plt)
Tetrad (.hdf5, .h5)
UNIC (.h5)
VASP CHGCA (.CHG)
VASP OUT (.OUT)
VASP POSTCAR (.POS)
VPIC (.vpc)
VRML (.wrl)
Velodyne (.vld, .rst)
VizSchema (.h5, .vsh5)
Wavefront Polygonal Data
(.obj)
WindBlade (.wind)
XDMF and hdf5 (.xmf, .xdmf)
XMol Molecule

Load disk_out_ref.ex2
1. Open the file disk_out_ref.ex2.
2. Load all data variables.




3. Click
Data Representation
Toggle Color
Legend
Mapped
Variable
Representation
Vector
Component
Edit Colors
Reset Scalar
Range
Geometry Representations
Points Wireframe Surface Surface
with Edges
Volume
Calculator
Contour
Clip
Slice
Threshold
Extract Subset
Common Filters
Glyph
Stream Tracer
Warp (vector)
Group Datasets
Extract Level
Filters Menu
Quick Launch

Used for searching
for filters by name
Keyboard shortcut
Ctrl-space for
Windows & Linux
Alt-space for Mac
Apply a Filter
1. Make sure that disk_out_ref.ex2 is
selected in the pipeline browser.
2. Select the contour filter.
Apply a Filter
3. Change parameters to create an
isosurface at Temp = 400K.
Change to Temp
Change to 400
Apply a Filter
1. Make sure that disk_out_ref.ex2 is
selected in the pipeline browser.
2. Select the contour filter.
3. Change parameters to create an
isosurface at Temp = 400K.
4.
Reset ParaView
Multiview
Multiview
1. Open disk_out_ref.ex2. Load all
variables.
2. Add clip filter.
3. Uncheck Show Plane.
4.
5. Color surface by Pres.
Multiview
6. Split the view horizontally.
7. Make Clip1 visible.
8. Color surface by Temp.
Multiview
6. Split the view horizontally.
7. Make Clip1 visible.
8. Color surface by Temp.
9. Right-click view background, Link
Camera
10.Click other view.
Multiview
6. Split the view horizontally.
7. Make Clip1 visible.
8. Color surface by Temp.
9. Right-click view background, Link
Camera
10.Click other view.
11.Click
Modifying Views
Modifying Views
Reset ParaView
Streamlines
1. Open disk_out_ref.ex2. Load all
variables.
2. Add stream tracer.
3.
Streamlines
1. Open disk_out_ref.ex2. Load all
variables.
2. Add stream tracer.
3.
4. From the quick launch, select Tube
5.
Getting Fancy
1. Select StreamTracer1.
2. Add glyph filter.
3. Change Vectors to V.
4. Change Glyph Type to Cone.
5.

Getting Answers
Where is the air moving the fastest?
Near the disk or away from it? At the
center of the disk or near its edges?
Which way is the plate spinning?
At the surface of the disk, is air moving
toward the center or away from it?
Reset ParaView
Common Data Analysis Filters
Extract Selection
Plot Global Variables Over Time
Plot Over Line
Plot Selection Over Time
Probe Location
Plotting
1. Open disk_out_ref.ex2. Load all
variables.
2. Clip, uncheck, ,
3. Select disk_out_ref.ex2
4. Add Plot Over Line filter.
3D Widgets
Plotting
1. Open disk_out_ref.ex2. Load all
variables.
2. Clip, uncheck, ,
3. Select disk_out_ref.ex2
4. Add Plot Over Line filter.
5. Once line is satisfactorily located,
Interacting with Plots
Left, middle, right buttons to pan, zoom.
Mouse wheel to zoom.
Reset view to plot ranges.
Plots are Views
Move them like Views.
Save screenshots.
Adjusting Plots
1. In Display tab, turn off all variables
except Temp and Pres.
2. Select Pres in the Display tab.
3. Change Chart Axis to Bottom Right.
4. Verify the relationship between
temperature and pressure.
Reset ParaView
Loading Data with Time
1. Open the file can.ex2.
2. Select all variables.
3.
4.
5.
Animation Toolbar
First
Frame
Previous
Frame
Play
Next
Frame
Last
Frame
Loop
Animation
Current
Time
Current
Time Step
Save Screenshot/Animation
1. Choose File Save Screenshot
2. Complete the following dialogs to
save an image.
3. Chose File Save Animation
4. Complete the following dialogs to
save a movie.
Query-Based Selection
1. Open can.ex2. All variables.
2. Go to last time step.
3. Edit Find Data.
4. Top combo box: find Cells.
5. Next row: EQPS, is >=, and 1.5.
6. Click Run Selection Query.
Query-Based Selection
Brush Selection
Surface Cell Selection (shortcut: s)
Surface Point Selection
Through Cell Selection
Through Point Selection
Select Points (polygon)
Select Cells (polygon)
Block Selection (shortcut: b)

Frustum vs. Id Selections
1. Make a Select Cells Through
2. Turn on Show Frustum in Find
Data. Rotate 3D view.
3. Play
4. Hit Freeze Selection in Find Data.
5. Play
Adding Labels
1. Select a few cells (not too many).
2. Open Find Data.
3. In the Labels chooser, select EQPS.
Plot Selection Over Time
1. Select a single cell where something
interesting happens.
2. From the quick launch, select Plot
Selection Over Time
3.
4. In Display panel, select different
blocks to plot.
Reset ParaView
Extracting a Selection
1. Open can.ex2. Load all variables.
2. Perform a sizeable selection.
3. Add the extract selection filter.
4.
Adding Text Annotation
1. If you have fell behind, you can reset
ParaView and reload can.ex2.
2. Sources Text
3. Type a message in text edit box
4.
Text Position
Annotate Time
1. Sources Annotate Time
2.
Python Trace Demo

Visualizing Large Models
Golevka Asteroid vs.
10 Megaton Explosion
CTH shock physics, over 1 billion cells
Polar Vortex Breakdown
SEAM Climate Modeling, 1 billion cells
(500 million cells visualized).
Objects-in-Crosswind Fire
Coupled SIERRA/Fuego/Syrinx/Calore,
10 million unstructured hexahedra
Large Scale AMR
Wing with Unsteady Crossflow
from Synthetic Jet (3.3B tet)
ParaView Architecture
Three tier
Data Server
Render Server
Client
Standalone
Client
Data
Server
Render
Server
Client-Server
Client
Data
Server
Render
Server
Client-Render Server-Data Server
Client
Data
Server
Render
Server
Requirements for
Installing ParaView Server
C++
CMake (www.cmake.org)
MPI
OpenGL (or Mesa3D www.mesa3d.org)
Qt 4.6 (optional)
Python (optional)
http://www.paraview.org/Wiki/Setting_up_a_ParaView_Server#Compiling
Connecting to a ParaView Server
http://www.paraview.org/Wiki/Setting_up_a_ParaView_Server#Running_the_Server
Data Parallel Pipelines
Duplicate pipelines run independently
on different partitions of data.
Data Parallel Pipelines
Duplicate pipelines run independently
on different partitions of data.
Data Parallel Pipelines
Some operations will work regardless.
Example: Clipping.
Data Parallel Pipelines
Some operations will work regardless.
Example: Clipping.
Data Parallel Pipelines
Some operations will work regardless.
Example: Clipping.
Data Parallel Pipelines
Some operations will have problems.
Example: External Faces
Data Parallel Pipelines
Some operations will have problems.
Example: External Faces
Data Parallel Pipelines
Ghost cells can solve most of these
problems.
Data Parallel Pipelines
Ghost cells can solve most of these
problems.
Data Partitioning
Partitions should be load balanced and
spatially coherent.
Data Partitioning
Partitions should be load balanced and
spatially coherent.
Data Partitioning
Partitions should be load balanced and
spatially coherent.
Load Balancing/Ghost Cells
Automatic for Structured Meshes.
Partitioning/ghost cells for unstructured
is manual.
Use the D3 filter for unstructured
(Filters Alphabetical D3)
Job Size Rules of Thumb
Structured Data
Try for max 20 M cell/processor.
Shoot for 5 10 M cell/processor.
Unstructured Data
Try for max 1 M cell/processor.
Shoot for 250 500 K cell/processor.
Avoiding Data Explosion
Pipeline may cause data to be copied,
created, converted.
This advice only for dealing with very
large amounts of data.
Remaining available memory is low.
Topology Changing, No Reduction
Append Datasets
Append Geometry
Clean
Clean to Grid
Connectivity
D3
Delaunay 2D/3D
Extract Edges
Linear Extrusion
Loop Subdivision
Reflect
Rotational Extrusion
Shrink
Smooth
Subdivide
Tessellate
Tetrahedralize
Triangle Strips
Triangulate
Topology Changing,
Moderate Reduction
Clip
Decimate
Extract Cells by
Region
Extract Selection
Quadric Clustering
Threshold
Similar: Extract Subset
Topology Changing,
Dimension Reduction
Cell Centers
Contour
Extract CTH
Fragments
Extract CTH Parts
Extract Surface
Feature Edges
Mask Points
Outline (curvilinear)
Slice
Stream Tracer
Adds Field Data
Block Scalars
Calculator
Cell Data to Point Data
Compute Derivatives
Curvature
Elevation
Generate Ids
Gen. Surface Normals
Gradient
Level Scalars
Median
Mesh Quality
Octree Depth Limit
Octree Depth Scalars
Point Data to Cell Data
Process Id Scalars
Random Vectors
Resample with dataset
Surface Flow
Surface Vectors
Texture Map to
Transform
Warp (scalar)
Warp (vector)
Total Shallow Copy or
Output Independent of Input
Annotate Time
Append Attributes
Extract Block
Extract Datasets
Extract Level
Glyph
Group Datasets
Histogram
Integrate Variables
Normal Glyphs
Outline
Outline Corners
Plot Global Variables
Over Time
Plot Over Line
Plot Selection Over
Time
Probe Location
Temporal Shift Scale
Temporal Snap-to-
Time-Steps
Temporal Statistics
Special Cases
Temporal Filters
Temporal Interpolator
Particle Tracer
Temporal Cache
Programmable Filter
Culling Data
Reduce dimensionality early.
Contour and slice see inside volumes.
Prefer data reduction over extraction.
Slice instead of Clip.
Contour instead of Threshold.
Only extract when reducing an order of
magnitude or more.
Can still run into troubles.
Culling Data
Experiment with subsampled data.
Extract Subset
Use caution.
Subsampled data may be lacking.
Use full data to draw final conclusions.
Memory Inspector
Rendering Modes
Still Render
Full detail render.
Interactive Render
Sacrifices detail for speed.
Provides quick rendering rate.
Used when interacting with 3D view.
Level of Detail (LOD)
Geometric decimation
Used only with Interactive Render
Original Data Divisions: 50x50x50 Divisions: 10x10x10
3D Rendering Parameters
Edit Settings, Render View General
Parallel Rendering
Parallel Rendering
Tiled Displays
Image Size LOD
ParaViews parallel rendering overhead
proportional to image size.
Can use smaller images for interactive
rendering.
Original Data Subsample Rate: 2 pixels Subsample Rate: 4 pixels Subsample Rate: 8 pixels
Parallel Rendering Parameters
Edit Settings, Render View Server
Parameters for Large Data
Use Immediate Mode Rendering off for
GPU, on for CPU.
Try LOD Threshold off.
Also try LOD Resolution 10x10x10.
Always have remote rendering on.
Turn on subsampling.
Image Compression on.
Parameters for Low Bandwidth
Try larger subsampling rates.
Try Zlib compression and fewer bits.

Parameters for High Latency
Turn up Remote Render Threshold.
Turn on Use outline for LOD rendering if
necessary.
Play with the LOD Threshold and LOD
Resolution to control geometry sent to
client.
Try turning on Lock Interactive Render.
Thanks for Attending
Please Fill Out the Survey
http://bit.ly/sc13-tut-sh03
Further Reading
The ParaView Users Guide
http://paraview.org/Wiki/ParaView/Users_Guide/Table_Of_C
ontents
http://www.paraview.org/Wiki/ParaView
http://www.paraview.org/Wiki/Setting_up_a_ParaView_Server
Further Reading
Visualization and Customization
Will Schroeder, Ken Martin, and Bill
Lorensen. The Visualization Toolkit.
Kitware, Inc., fourth edition, 2006.
Kitware Inc. The VTK Users Guide.
Kitware, Inc., 2006.
J asmin Blanchette and Mark
Summerfield. C++ GUI Programming
with Qt 4. Prentice Hall, 2006.
Further Reading
Parallel VTK Topics
J ames Ahrens, Charles Law, Will Schroeder, Ken
Martin, and Michael Papka. A Parallel Approach for
Efficiently Visualizing Extremely Large, Time-Varying
Datasets. Technical Report #LAUR-00-1620, Los
Alamos National Laboratory, 2000.
J ames Ahrens, Kristi Brislawn, Ken Martin, Berk
Geveci, C. Charles Law, and Michael Papka. Large-
Scale Data Visualization Using Parallel Data
Streaming. IEEE Computer Graphics and
Applications, 21(4): 3441, J uly/August 2001.
Andy Cedilnik, Berk Geveci, Kenneth Moreland,
J ames Ahrens, and J ean Farve. Remote Large
Data Visualization in the ParaView Framework.
Eurographics Parallel Graphics and Visualization
2006, pg. 163170, May 2006.
Further Reading
Advanced Pipeline Execution
J ames P. Ahrens, Nehal Desai, Patrick S. McCormic,
Ken Martin, and J onathan Woodring. A Modular,
Extensible Visualization System Architecture for
Culled, Prioritized Data Streaming. Visualization
and Data Analysis 2007, Proceedings of SPIE-IS&T
Electronic Imaging, pg 64950I-112, J anuary 2007.
J ohn Biddiscombe, Berk Geveci, Ken Martin,
Kenneth Moreland, and David Thompson. Time
Dependent Processing in a Parallel Pipeline
Architecture. IEEE Visualization 2007. October
2007.
Further Reading
Parallel Rendering
Kenneth Moreland, Brian Wylie, and Constantine
Pavlakos. Sort-Last Parallel Rendering for Viewing
Extremely Large Data Sets on Tile Displays.
Proceedings of IEEE 2001 Symposium on Parallel
and Large-Data Visualization and Graphics, pg. 85
92, October 2001.
Kenneth Moreland and David Thompson. From
Cluster to Wall with VTK. Proceddings of IEEE 2003
Symposium on Parallel and Large-Data Visualization
and Graphics, pg. 2531, October 2003.
Kenneth Moreland, Lisa Avila, and Lee Ann Fisk.
Parallel Unstructured Volume Rendering in
ParaView. Visualization and Data Analysis 2007,
Proceedings of SPIE-IS&T Electronic Imaging, pg.
64950F-112, J anuary 2007.