Advanced Imaging


Advanced Imaging Magazine

Updated: July 8th, 2008 05:26 PM CDT

Between a Rock and a Hard Place

Mercury Computer Systems
This display shows a variety of different types of data and visualizations typical of oil exploration and production research.
Mercury Computer Systems
Mercury's data management strategy for the very large volumes of data in oil E&P. The data is processed and visualized in 3D by the GPU.
Mercury Computer Systems
An advanced visualization blends amplitude and velocity into a single image by use of a shader, a specific set of software instructions.
Mercury Computer Systems
A small cluster of PCs can be used to show scalability (ability to handle excessive amounts of data).

By Barry Hochfelder

3D visualization has become entwined with computation to help sift through the data. Oil companies work under tremendous pressure to find new sources within the fields and how to get it out most efficiently. In many cases there are multiple data sets, including raw seismic data, along with dozens of attributes and mounds of physical data. The idea is to make the information more effectively visual for them.

The technology, the basic principle of visualizing polygonal and volume data has been around for years. In the past, the E&P industry used large machines for both computation and rendering. Heck explains that, as the economics of the PC architecture overtook the big machines, it didn't seem that the capabilities of a single machine would be sufficient. "The industry turned to clusters of PCs as a solution," he says."

While they were primarily used for computational tasks, they were used in a limited manner for visualization. Advances in data management, computing and rendering, have made a single machine viable for visualization of E&P data.

The solution, Heck says, is advanced middleware for visualization, which frees application developers to focus on turning their domain-specific knowledge into complete solutions.

"Sixty-four bit operating systems have enabled much larger system memory, but both system memory and texture memory on the graphics processing unit remain scarce resources compared to the size of the data sets," he says. "An effective solution using hierarchical multi-resolution bricking now is available in visualization tool kits—middleware.

Subscribe to our RSS Feeds