• <More on Intel.com

Shell Drills Down on Improvements in High-Performance Computing

We are sorry, This PDF is available in download format only

High-performance computing (HPC) is mission-critical for Shell. Having more HPC capacity helps us explore for and find hydrocarbons in increasingly more complex and difficult geologic environments, extract resources efficiently, and help meet the world’s energy needs while minimizing environmental impact. Shell’s demand for HPC capacity is increasing dramatically. For our upstream exploration businesses, newer seismic acquisition techniques are generating larger and larger seismic datasets, which alongside new seismic algorithms are creating more and more demand for HPC capacity. In addition, dramatically more complex and larger reservoir models, combined with new reservoir simulator technologies, are increasing HPC demand. Shell’s research and development (R&D) programs and other new emerging HPC opportunities also require next-generation HPC capabilities as well. A key driver for Shell is the desire to deliver the required HPC capacity at the best-possible total cost of ownership (TCO). We deliver as much capacity as possible from Shell’s global data centers in Houston and Amsterdam, and in general try to concentrate capacity in fewer locations. This approach helps us create scale benefits in cost and service levels and will help us transition to cloud-based solutions as they mature and become more affordable in the future. Because of HPC’s importance, Shell established an HPC Center of Excellence (CoE) in 2009, and set a goal of hundredfold improvements in application price/ performance. Today, through rapid adoption of the latest technologies, an ongoing focus on software optimization, and close collaboration within Shell and with external leaders, we are well on our way to achieving that goal. In doing so, we are delivering technical successes, enabling significant results, and providing competitive advantages.