Database Reference
In-Depth Information
IPR_C vs IPR_D vs ts
ts
IPR_C
IPR_D
(b)
Figure 9.15
(
Continued
)
technology with visual data analysis technology to tackle problems of scale in
a manner that virtually guarantees scientific insight and knowledge discovery.
9.6 Summary and Conclusion
Visualization, which is the transformation of abstract data into images, plays
a central role in virtually all fields of scientific endeavor. It is an indispensable
part of hypothesis testing and knowledge discovery. Like most other fields dis-
cussed in this topic, visualization faces substantial scientific data management
challenges that are the result of growth in size and complexity of data being
produced by simulations and collected from experiments. Our objective in this
chapter has been to reveal some of the scientific data management issues, chal-
lenges, and solutions that are somewhat unique to the field of visualization.
Production visualization applications, namely those that run on large-scale
parallel machines, can derive great benefit from close attention to scientific
data management issues. Staple operations, like slicing and isosurfacing, can
be vastly accelerated by taking advantage of metadata. In some cases, the sim-
ulation or experiment produces such metadata as part of the data production
process. In other cases, we must generate that data ourselves. A significant
fraction of the code in these production visualization applications is dedicated
