Top Banner
Large-Scale Visual Analysis Chris Johnson Scientific Computing and Imaging Institute University of Utah
45

Large-Scale Visual Analysis

Feb 26, 2022

Download

Documents

dariahiddleston
Welcome message from author
This document is posted to help you gain knowledge. Please leave a comment to let me know what you think about it! Share it to your friends and learn new things together.
Transcript
Page 1: Large-Scale Visual Analysis

Large-Scale Visual Analysis

Chris JohnsonScientific Computing and Imaging Institute

University of Utah

Page 2: Large-Scale Visual Analysis

History of Computer Graphics in Utah

Page 3: Large-Scale Visual Analysis

SCI Institute Faculty

Page 4: Large-Scale Visual Analysis

Research Cores

Visualization

Computing ImageAnalysis

Scientific DataManagement

Page 5: Large-Scale Visual Analysis

Visualization

Computing ImageAnalysis

Software Development

Page 6: Large-Scale Visual Analysis

Centers We DirectNIH/NIGMS Center for IntegrativeBiomedical Computing

Center for Extreme Data Management,Analysis, and Visualization

CEDMAV

Utah Center for Neuroimage Analysis

Page 7: Large-Scale Visual Analysis

National Centers We are Affiliated With

CDC Decision-Support for Infectious Disease

Epidemiology

NIH NAMIC

IAMCSInstitute for Applied Mathematicsand Computational Science

Scalable Data Management, Analysisand Visualization

Center for Exascale Simulationof Combustion in Turbulence

Page 8: Large-Scale Visual Analysis

Sources: Lesk, Berkeley SIMS, Landauer, EMC, TechCrunch, Smart Planet

0.1

1

10

100

1000

1999 2001 2003 2005 2007 2011

all disk storage

all digital info

new digital info/yr

all human documents in 40k Yrs

all spoken words in all lives

amount human minds can store in 1yr

Exab

ytes

(10

)18

912

2012

Every two days we create as much data as we did from the beginning of mankind until 2003!

Page 9: Large-Scale Visual Analysis

How Much is an Exabyte?

1 Exabyte = 1000 Petabytes = could hold approximately500,000,000,000,000 pages of standard printed text

It takes one tree to produce 94,200 pages of a book

Thus it will take 530,785,562,327 trees to store an Exabyte of data

In 2005, there were 400,246,300,201 trees on Earth

We can store .75 Exabytes of data using all the trees on the entire planet.

Sources: http://www.whatsabyte.com/ and http://wiki.answers.com

How many trees does it take to print out an Exabyte?

Page 10: Large-Scale Visual Analysis

Scientific Computing and Imaging Institute, University of Utah

Brain Information Bandwidth

amount'we're'actually'aware'of'(0.7%)

sight

touch

hearing/smell

tastehard'diskUSB'keycomputer'network

Image'courtesy'Tor'Norretranders

pocket'calculator

0.

Page 11: Large-Scale Visual Analysis

Scientific Computing and Imaging Institute, University of Utah

Feynman Diagrams

Feynman: “What I am really try to do is bring birth to clarity, which is really a half-assedly thought-out-pictorial semi-vision thing. I would see the jiggle-jiggle-jiggle or the wiggle of the path. Even now when I talk about the influence functional, I see the coupling and I take this turn - like as if there was a big bag of stuff - and try to collect it in away and to push it. It's all visual. It's hard to explain.”

James Gleick, The Life and Science of Richard Feynman, Vintage Books, New York, 1992.

Page 12: Large-Scale Visual Analysis

New Visual Analysis Techniques

Page 13: Large-Scale Visual Analysis
Page 14: Large-Scale Visual Analysis
Page 15: Large-Scale Visual Analysis

Volume Rendering

Page 16: Large-Scale Visual Analysis

Scientific Computing and Imaging Institute, University of Utah

Volume Rendering

Maximum Intensity Projection (MIP) Full Volume Rendering

Page 17: Large-Scale Visual Analysis

Scientific Computing and Imaging Institute, University of Utah

Multi-Dimensional Transfer Function

0

+

-RGB ( , ,

)

Page 18: Large-Scale Visual Analysis

Scientific Computing and Imaging Institute, University of Utah

Volume Rendering

enamel /background

dentin / background dentin / enamel dentin / pulp

1D: not possible2D: specificity not as good

Page 19: Large-Scale Visual Analysis

Scientific Computing and Imaging Institute, University of Utah

Volume Visualization

Page 20: Large-Scale Visual Analysis

Scientific Computing and Imaging Institute, University of Utah

NIH Visible Male

Page 21: Large-Scale Visual Analysis

Scientific Computing and Imaging Institute, University of Utah

Visible Human - High Resolution

Page 22: Large-Scale Visual Analysis

The Need for High Resolution Visualization

Lower Resolution High Resolution

“…the data show for the first time how detailed transport and chemistry effects can influence the mixing of reactive scalars. It may be advantageous to incorporate these effects within molecular mixing models. It is worth noting that at present it is impossible to obtain this type of information any other way than by using the type of highly resolved simulation performed here.”Jacqueline Chen, Sandia National Laboratories

Page 23: Large-Scale Visual Analysis
Page 24: Large-Scale Visual Analysis

University of Utah

§ Non-­‐premixed  DNS  combustion  (J.  Chen,  SNL):    Analysis  of  the  time  evolution  of  extinction  and  reignition  regions  for  the  designof  better  fuels

Topological  Analysis  of  Massive  Combustion  Simulations

Page 25: Large-Scale Visual Analysis

University of Utah

New Parallel Topological Computations Achieve High Performance at Scale

Pure Computation

Computation + I/O

Page 26: Large-Scale Visual Analysis

Visualization of 10D Combustion Simulation of Jet CO/H2-Air Flames

10 dimensional data set describing the heat release wrt. to various chemical species in a combustion simulation

Pure Fuel

Pure Oxidizer

Local Extinction

Page 27: Large-Scale Visual Analysis

Analysis of Combustion Simulations

Page 28: Large-Scale Visual Analysis

Scientific Computing and Imaging Institute, University of Utah

Michelangelos David

Page 29: Large-Scale Visual Analysis

Scientific Computing and Imaging Institute, University of Utah

Michelangelos David - Part 2

Page 30: Large-Scale Visual Analysis

University of Utah

ViSUS  Framework  for  Scalable  Data  

Page 31: Large-Scale Visual Analysis

Scientific Computing and Imaging Institute, University of Utah

Large Scale Galaxy Simulation

Page 32: Large-Scale Visual Analysis

341 Sections90nm thick sections

~32GB/Section~1000 tiles/section

4096x4096 pixels/tile2.18 nm/Pixel

16.5 TB after processing

Page 33: Large-Scale Visual Analysis

Scientific Computing and Imaging Institute, University of Utah

Antony van Leeuwenhoek (1632-1723)

. . . my work, which I've done for a long time, was not pursued in order to gain the praise I now enjoy, but chiefly from a craving after knowledge, which I notice resides in me more than in most other men. And therewithal, whenever I found out anything remarkable, I have thought it my duty to put down my discovery on paper, so that all ingenious people might be informed thereof.

Antony van Leeuwenhoek. Letter of June 12, 1716

Page 34: Large-Scale Visual Analysis

Scientific Computing and Imaging Institute, University of Utah

Page 35: Large-Scale Visual Analysis

Scientific Computing and Imaging Institute, University of Utah

Uncertainty Visualization

When is the last time you’ve seen an error bar in a 3D visualization?

Page 36: Large-Scale Visual Analysis

Scientific Computing and Imaging Institute, University of Utah

Uncertainty VisualizationSurfaces imply certainty

Page 37: Large-Scale Visual Analysis

Scientific Computing and Imaging Institute, University of Utah

Uncertainty VisualizationSurfaces imply certainty

Page 38: Large-Scale Visual Analysis

Scientific Computing and Imaging Institute, University of Utah

Uncertainty VisualizationSurfaces imply certainty

Page 39: Large-Scale Visual Analysis

Scientific Computing and Imaging Institute, University of Utah

Topological Uncertainty

 M. Otto, T. Germer, H.C. Hege, H. Theisel. Uncertain 2D Vector Field Topology. In CGF, 29(2), 2010.

Page 40: Large-Scale Visual Analysis

Scientific Computing and Imaging Institute, University of Utah

Visualizing Uncertainty

Fuzzy Sensitivity Confidence

Page 41: Large-Scale Visual Analysis

QuizLens: A Multi-lens approach for uncertainty exploration

• Global information important for qualitative evaluation & context

• Local information necessary for quantitative understanding

• Interchangeable lenses to explore various data characteristics

Probability Slice

Context High Level Focus Lens

Fuzzy Volume Uncertain Boundary

Page 42: Large-Scale Visual Analysis

The SCI Institute

Page 43: Large-Scale Visual Analysis

Productivity Machines

Page 44: Large-Scale Visual Analysis

Scientific Computing and Imaging Institute, University of Utah

Acknowledgements

NIH NIGMSDOE SciDAC and NETLNSF Utah Centers of ExcellenceKAUST

Page 45: Large-Scale Visual Analysis

Scientific Computing and Imaging Institute, University of Utah

More Information

www.sci.utah.edu

[email protected]