Toward Effective Visualization of Ultrascale TimeVarying Data

1
Toward Effective Visualization of Ultra-scale
Time-Varying Data

• Han-Wei Shen
• Associate Professor
• The Ohio State University

2
Applications

• Large Scale Time-Dependent Simulations
• Richtmyer-Meshkov Turbulent Simulation (LLNL)
• 2048x2048x1920 grid per time step (7.7 GB)
• Run 27,000 time steps
• Multi-terabytes output

LLNL IBM ASCI system
3
Applications

• Oak Ridge Terascale Supernova Initiative (TSI)
• 640x640x640 floats
• gt 1000 time steps
• Total size gt 1 TB
• NASAs turbo pump simulation
• Multi-zones
• Moving meshes
• 300 time steps
• Total size gt 100GB

ORNL TSI data
NASA turbo pump
4
Research Goals and Challenges

• Interactive data exploration
• Quick overview, detail on demand
• Feature enhancement and tracking
• Display the invisible
• Understand the evolution of salient features over
  time
• Challenges
• managing, indexing, and processing of data

5
Research Focuses

• Multi-resolution data management schemes
• Acceleration Techniques
• Efficient data indexing
• Coherence exploitation
• Effective data culling
• Parallel and distributed processing
• Feature tracking and enhancement
• Visual representation
• Geometric tracking

6
Bricking and Multi-resolution

• Bricking subdivide the volume into mutiple
  blocks

7
Bricking and Multi-resolution

• Create a multi-resolution representation for each
  block

8
Spatial Data Hierarchy

• Combining octree with multi-res transform

bricks
9
Temporal Data Hierarchy?

• Option1 - Multiple Octrees

t 0 t 1
t 2
10
Temporal Data Hierarchy?

• Option 2 Treat time as another dimension a
  single 4D tree (16 tree)

11
Time-Space Partition (TSP) Tree(Two Level
Hierarchical Subdivision)

• First level spatial subdivision
  

bricks
Shallow Complete Octree
12
Time-Space Partition (TSP) Tree(Two Level
Hierarchical Subdivision)

• Second level temporal subdivision
  

4 time steps
13
Spatio-Temporal Data Encoding

• Wavelet Transform (DWT)

3D wavelet transform
1D WT
14
Spatio-Temporal Data Indexing

• Time-Space Partitioning (TSP) Trees
  

15
Tree Traversal and Rendering
16
Image Compositing
Front-to-back
17
Rendering Performance

• The cached partial images can be re-used
  for the nodes that have high temporal coherence
  

18
Time-Varying Volume Rendering
Error 0
11.2 speedup
19
I/O Efficiency
Shock wave 1024 x 128 x 128 , 40 time
steps Minimum brick size 32 x 32 x 32 Temporal
error tolerance 0.02
20
Time-Space Partition (TSP) Tree

• More cohesively integrate the temporal and
  spatial information into a single hierarchical
  data structure
• Exploit both temporal and spatial coherence -
  Octree becomes a special case of the TSP tree

21
Analyzing Time-varying Features

• Animation might not be sufficient

22
Strategy 1 Tracking individual components
23
Strategy 2 High Dimensional Visualization

• Chronovolumes

24
Tracking Time-Varying Isosurface

• Two main goals
• Identify correspondence
• Detect important evolution events and critical
  time steps

?
25
Evolutionary Events
26
Tracking Correspondence

• Wang and Silvers assumption - Corresponding
  features in adjacent time steps overlap with each
  other

27
Tracking Correspondence

• A common assumption - Corresponding features in
  adjacent time steps overlap with each other

t 0 t 1
28
Previous Approach

• Algorithm
• Extract the complete set of isosurfaces
• Overlap test
• Overlapping features are identified and the
  number of intersecting nodes is calculated.
• Best matching test
• Find the best match among features.

29
Challenges

• Exhaust search is expensive
• Solution A local tracking
• The user selects a local
• feature of interest and start
• tracking
• Extract high dimensional (4D) isosurfaces

30
2D Example

• 2D time-varying isocontours

T 2
T 1
T 0
31
2D Example

• Extract 3D isosurface and then slice back

T 2
T 1
T 0
32
2D Example

• Extract 3D isosurface and then slice back

T 2
T 1
T 0
33
4D Isosurface

• 3D time-varying 4D
• Extract isosurfaces from 4D hypercubes
• Use 4D maching cubes table (Bhaniramka02)
• Slice the tetrahedra to get the surface at the
  desired time step

(x,y,z,t)
34
Algorithm

• To track an isosurface component
• User chooses a local component at t
• Propagate 4D isosurface from the seed
• Slice the 4D isosurface at t1
• Continue to t2 if desired

35
Detect critical time steps for isosurface tracking

• A 4D isocontour component is a tetrahedral mesh
  embedded in four dimensional space. We can treat
  the 4D mesh as a normal 3D mesh, with the time
  values as the scalar values defined over the
  tetrahedron vertices.
• The critical points of this mesh indicate when
  and where the topology of the isosurface will
  change.
• Local minimum Creation
• Local maximum Dissipation
• Saddle Amalgamation/Bifurcati
  on
• Regular vertex Continuation

36
Color the components
37
Color the components
38
Critical Time Steps
39
Chronovolumes

• A Direct Rendering Technique for Visualizing
  Time-Varying Data

(Jonathan Woodring and Han-Wei Shen 2003)
40
Main Idea

• Render data at different time steps to a single
  image
• Establish correspondences between features
• Compare shapes and sizes of features in time
• Reason about the positions of the features
• Reveal temporal trend

41
Early Work
Chronophtography (Marey, 1830-1904)
Nude descending a staircase Duchamp, 1912
42
Chronovolumes

• 4D rendering idea
• Integration through time
• Integration functions

43
4D Rendering

• Direct visualization of 4D data
• Project the 4D data into a visualizable lower
  dimensional space (2D images)

2D -gt 1D
3D -gt 2D
44
4D Rendering

• 4D to 2D projection?
• Need to preserve the relationships between
  different objects in (3D) space and also reveal
  their relationship in time

45
Integration Through Time

• 4D to 3D projection (chronovolume)
• Regular volume rendering to visualize
  chronovolumes

chronovolume
46
Integration Function

• Vc F (Vt, V t1, V t2, V t3, , V tn-1)
• No so called correct integration the design
  of F depends on the visualization need

???
47
Alpha Compositing

• Commonly used in 3D volume rendering

D
0
C
2D Image
48
Alpha Compositing (2)

• Adopt the model to time integration

post-classified (color) volume
49
Transfer Function

• Color and opacity function
• Modulate by time stamp and data

Alpha function example
a
a

0.2
0.7
t
v
3 8
6
50
Alpha Compositing Example
10 time steps
3 time steps
51
Additive Colors

• Show how features overlap

T
T


C c(s(x(t)) dt
t4
0
t3
t2
t1
t
52
Additive Color Example
Alpha Compositing Additive Color
53
Additive Color Example
Alpha Compositing
Additive Colors
54
Additive Color Example
Alpha Compositing
Additive Colors
55
Min/Max Intensity

• Detect the hot spot

F(V i) t such that V t gt Vi for any i
lt

• Show which time step has the highest
• (lowest) value, and also what that
• value is.

56
Maximum Intensity Example
Additive Colors Maximum
Intensity
57
Maximum Intensity Examples
Alpha Compositing Maximum
Intensity