SC05

1
Desktop Techniques for the Exploration of
Terascale Sized Turbulence Data Sets

• John Clyne
• Scientific Computing Division
• National Center for Atmospheric Research
• Boulder, CO USA

2
Numerical models that can currently be run on
typical supercomputing platforms produce data in
amounts that make storage expensive, movement
cumbersome, visualization difficult, and detailed
analysis impossible.  The result is a
significantly reduced scientific return from the
nation's largest computational efforts.
We can now compute more data than we know how to
analyze!!!
3
A sampling of various technology performance
curves

• Not all technologies advance at same rate!!!

4
(No Transcript)
5
(No Transcript)
6
Growth of individual NCAR simulation data sets
Representative data sets from climate,
turbulence, and weather disciplines
7
Climate simulation grid resolutions
8
Example Compressible plume dynamics

• 504x504x2048
• 5 variables (u,v,w,rho,temp)
• 500 time steps saved
• 9TBs storage (4GBs/variable/timestep)
• Six months compute time required on 112 IBM SP
  RS/6000 processors
• Three months for post-processing
• Data may be analyzed for several years

M. Rast, 2004. Images courtesy of Joseph Mendoza,
NCAR/SCD
9
Workflow in computational science
Simulation
Analysis Visualization
Post Processing
Storage
Storage
10
What is meant by interactive computing?

• Definition A system is interactive if the time
  between a user event and the response to that
  event is short enough maintain my full attention
• If the response time is
• 1-5 seconds Im engaged
• 5-60 seconds Im reading email
• 1-3 minutes Ive forgotten what I was trying to
  do
• gt 3 minutes Ive given up!

11
IO wait times for high resolution simulations
Resolution MBs per variable Scalar variable wait time Vector variable wait time
1283 8 0.1 0.3
2563 67 0.7 2.1
5123 537 5.0 15.0
10243 4295 43.0 130.0

• Assumptions
• Single precision
• 100 MB/sec bandwidth
• No contention

12
Visualization and Analysis Platform for oceanic,
atmospheric, and solar Research (VAPoR)

• Key components
• Domain specific application focus numerically
  simulated turbulence
• Quantitative capabilities to support scientific
  data analysis
• Integrate visualization into analysis process,
  interactively steering the analysis while
  enhancing data understanding
• Employ multiresolution data representation as a
  data reduction technique

Combination of visualization with multiresolution
data representation that provide sufficient data
reduction to enable interactive work
This work is funded in part through a U.S.
National Science Foundation, Information
Technology Research program grant
13
Enabling speed/quality tradeoffs with
multiresolution data representation

• 2D Example Texture MIP Mapping

1
14
Wavelet Transforms for 3D Multiresolution data
representation

• Permit hierarchical data representation
• Invertible and lossless (subject to floating
  point round off errors)
• Numerically efficient forward and inverse
  transform
• No additional storage cost!!!

15
Compressible Convection
M. Rast, 2002
1283
5123
16
Compressible plume
Compressible plume data set shown at native and
progressively coarser resolutions
504x504x2048 Full
252x252x1024 1/8
126x126x512 1/64
63x63x256 1/512
Resolution Problem size
17
Rendering timings
5123 Compressible Convection
5042x2048 Compressible Plume
SGI Octane2, 1x600MHz R14k
SGI Origin, 10x600MHz R14k
Reduced resolution affords responsive interaction
while preserving all but finest features
18
Derived quantities
Derived quantities produced from the simulations
field variables as a post-process

• ?? Avogadros number
• me electron mass
• k Boltzmanns constant
• h Plancks constant

• p pressure
• ? density
• T temperature
• ?? ionization potential

19
Calculation timings for derived quantities
SGI Origin, 10x600MHz R14k
Note 1/2th resolution is 1/8th problem size, etc
Deriving new quantities on interactive time
scales only possible with data reduction
20
Integrated visualization and analysis on
interactively selected subdomains
Mach number of the vertical velocity.
Efficient analysis requires rapid calculation and
visualization of unanticipated derived
quantities. This can be facilitated by a
combination of subdomain selection and resolution
reduction.
Vertical vorticity of the flow.
21
A test of multiresolution analysis Force
balance in supersonic downflows
Resolution
Full
Half
Subdomain selection and reduced resolution
together yield data reduction by a factor of
128!!!
Sites of supersonic downflow are also those of
very high vertical vorticity. The cores of the
vortex tubes are evacuated, with centripetal
acceleration balancing that due to the inward
directed pressure gradient. Buoyancy forces are
maximum on the tube periphery due to mass flux
convergence. The same interpretation results
from analysis at half resolution.
22
Future???
Original
201 Lossy Compression
23

• Live VAPOR demonstrations, SGI Theatre (booth
  602)
• Wednesday, 1130am
• Thursday, 330pm
• VAPOR URL
• http//www.scd.ucar.edu/hss/dasg/software/vapor

Questions???
24

• Inadequate IO bandwidth is but one impediment to
  interactive analysis and visualization.
• Others impediments include
• Insufficient capacity of high-speed storage
• Reliance on un-optimized, serial applications
• Mismatch between simulation and analysis
  computing resources

25
NCAR Science
More than just the atmosphere from the earths
oceans to the solar interior
Turbulence
The Sun
Space Weather
Atmospheric Chemistry
Climate
Weather