FIGURE 6.3 Data sets used for IC tests. Integral curves through the magnetic field of a GenASiS
supernova simulation (left), data courtesy of Anthony Mezzacappa (ORNL). Integral curves through
the magnetic field of a NIMROD fusion simulation (right), data courtesy of Scott Kruger (Tech-X
Corporation).
TABLE 6.1 Performance Benchmarks
ermal Hydraulics Hydrodynamics Combustion
Strong scaling, 2048
×2048× 2048× 1
timestep = 98GB data,
128K particles
Weak scaling, 2304
× 4096 × 4096 × 1
timestep = 432GB data,
16K to 128K particles
Weak scaling, 1408
× 1080 × 1100 × 32
timesteps = 608GB data,
512 to 16K particles
(a) (b) (c)
FIGURE 7.3 The left images, (a) and (b), show a comparison of two different parallel coordinate
renderings of a particle data set consisting of 256,463 data records and 7 variables using: (a) tradi-
tional line-based parallel coordinates and (b) high-resolution, histogram-based parallel coordinates
with 700 bins per data dimension. The histogram-based rendering reveals many more details when
displaying large numbers of data records. Image (c) shows the temporal histogram-based parallel
coordinates of two particle beams in a laser-plasma accelerator data set, at timesteps t = [14; 22].
Color is used to indicate the discrete timesteps. The two different beams can be readily identified in
x (second axis). Differences in the acceleration can be clearly seen in the momentum in the x direc-
tion, px (first axis). Image source: Rübel et al., 2008.
(a) (b) (c)
FIGURE 7.4 Applications of segmentation of query results. Image (a) displays the magnetic
confinement fusion visualization showing regions of high magnetic potential colored by their con-
nected component label. Image source: Wu et al., 2011. Image (b) shows the query selecting the
eye of a hurricane. Multivariate statistics-based segmentation reveals three distinct regions in
which the query’s joint distribution is dominated by the influence of pressure (blue), velocity (green),
and temperature (red). Image source: Gosink et al., 2011. Image (c) is the volume rendering of the
plasma density (gray), illustrating the wake of the laser in a plasma-based accelerator. The data
set contains approximately 229 w 10
6
particles per timestep. The particles of the two main beams,
automatically detected by the query-based analysis, are shown colored by their momentum in accel-
eration direction (px). Image source: Rübel et al., 2009.
(a)
0 200 400 600 800
0
2
4
6
8
10
12
14
16
x 10
5
time step
number of halo particles
(b)
FIGURE 7.6 (a) Particle density plot (gray) and selected halo-query particles (red) for timestep
20 of the simulation. (b) Plot showing the number of halo particles per timestep in a 50TB electron
linear particle accelerator data set. Image source: Chou et al., 2011..
!
"
"
#
$
!
"
"
#
$
(a)
(b)
(c)
!
"
"
#
$
FIGURE 7.5 Histograms showing the number of unsuccessful connection attempts (with radia-
tion excluded): (a) on ports 2000 to 65535 over a 42-week period, indicating high levels of activity
on port 5554 during the 7th week; (b) per source A octet during the 7th week on port 5554, indicating
suspicious activity from IPs with a 220 A octet; (c) per destination C octet scanned by seven suspi-
cious source hosts (color), indicating a clear scanning pattern. Image source: Bethel et al., 2006.
FIGURE 7.7 QDV of large 3D plasma-based particle acceleration data containing approximately
90 x 10
6
particles per timestep. On the left (a), parallel coordinates of timestep t = 12 showing: (1)
all particles above the base acceleration level of the plasma wave (Query: px > 2 w 10
9
) (gray) and
(2) a set of particles that form a compact beam in the first wake period following the laser pulse
(Query: (px > 4:856 w 10
10
) AND (x > 5:649 w 10
–4
))(red). On the right (b), volume rendering of the
plasma density illustrating the 3D structure of the plasma wave. The selected beam particles are
shown in addition in red. Image source: Rübel et al., 2008.
FIGURE 7.8 Visualization of the relative traces of a particle beam in a laser plasma accelerator.
The traces show the motion of the beam particles relative to the laser pulse. The xy-plane shows iso-
contours of the particle density at the timepoint when the beam reaches its peak energy, illustrating
the location of the beam within the plasma wave. The up-axis and color of the particle traces show
the particle momentum in the transverse direction, py. The image shows particles being injected
from the sides and oscillating while being accelerated. Image source: Rübel et al., 2009.
(a) (b)
(c) (d)
FIGURE 8.7 Direct volume rendering of an enstrophy field: original data (a), and shown in (b){(d),
respectively, are results after reduction of the number of expansion coefficients used in reconstruc-
tion by factors of
1
500
,
1
100
, and
1
10
.
..................Content has been hidden....................
You can't read the all page of ebook, please click here login for view all page.