Our Geoscientists Help Us Lead Industry in High Performance Computing Techniques
Access to high-quality seismic images of the subsurface is critical to successful exploration for oil and gas. Sound waves, which travel at different velocities as they pass through rocks, sand, salt, oil and gas, are used by geoscientists to illustrate and create a detailed understanding of subsurface structures and to identify possible oil and gas migration paths, the geometry and size of potential traps and the probable locations of hydrocarbons.
However, these images can be challenging to obtain, especially when salt domes or thousands of feet of rock bury the oil or gas. To improve the speed, quality and accuracy of subsurface images and interpretations — and drive business results — we rely on high performance computing.
In 1998 we pioneered the use of Linux computer CPU (central processing unit) clusters for seismic imaging. About a decade later, we spearheaded the petroleum industry’s conversion to doing seismic imaging on GPU (graphics processing unit) clusters.
We were the first energy company to use hundreds of GPUs from Nvidia, better known as the graphic brains inside computers with striking 3D graphics, in seismic applications. Clusters of computers containing GPUs offer better price-to-performance ratios than Linux clusters with PC CPUs. By using GPU clusters we improved the speed, quality and physical fidelity of our images.
GPUs are just one of numerous technology tools we use to improve the accuracy of subsurface images, making drilling safer and more reliable.
We continue to monitor and test GPUs and other accelerator technologies to meet the demands of computer intensive imaging paradigms including Reverse Time Migration (RTM), iterative and interactive model building and full waveform inversion (FWI). In addition to seismic imaging, our high performance computing is critical for everything from timely modeling and simulation of large-scale physical systems to processing, analysis, interpretation and visualization of huge seismic data volumes.
We are maximizing the potential of technology in the ways we use, combine and interact with it, incorporate it into our processes and integrate it into our culture and behaviors.