This post is an entry inThe World Isn’t Flat, It’s Parallel series running on nTersect, focused on the GPU’s importance and the future of parallel processing. Today, GPUs can operate faster and more cost-efficiently than CPUs in a range of increasingly important sectors, such as medicine, national security, natural resources and emergency services. For more information on GPUs and their applications, keep your eyes on The World Isn’t Flat, It’s Parallel.
For oil and gas companies, finding the precise location to drill can be worth hundreds of millions of dollars. Dig a few feet in the wrong direction, and over time the loss in production can mean significant lost revenue.
With stakes this high, it’s critical that companies involved in oil and gas exploration use the most advanced seismic imaging technology available. Increasingly, this compute-intensive work is being done with GPUs.
SeismicCity, a Houston-based leader in depth imaging services, uses an NVIDIA Tesla S1070 system to create incredibly precise images of geological data from deep inside the earth. The resulting models enable oil and gas companies to more quickly and accurately discover new oil and gas reserves.
SeismicCity made the change to GPUs a little over a year ago when it replaced its CPU-based data center. Not only did the switch produce a 10-fold performance boost, but it also allowed SeismicCity to increase the complexity of its algorithms, giving it a distinct competitive advantage. It developed a proprietary technique – Reverse Time Migration (RTM) – that is one of the most advanced seismic imaging technologies in the industry.
Technological innovations like RTM wouldn’t be possible without parallel computing. Earlier in this decade, traditional CPUs hit the “Power Wall” – they reached a point where it would take so much power to increase speed, it’s effectively impossible for them to get faster. And because CPUs are optimized for sequential tasks and sequential applications, they are limited when it comes to their use in large-scale scientific computations.
But GPUs are inherently parallel. In oil and gas exploration – as in other fields – computational tasks that are simply too resource intensive to process on CPU-based systems become possible with the massively parallel architecture of GPU computing.
For SeismicCity, the shift to parallel computing has meant it can continue to advance its algorithms without being held back by its hardware. For SeismicCity’s customers, it’s meant more timely discoveries of the oil and gas reserves so crucial to the future. It’s yet one more way that GPU computing is making acceleration breakthroughs that are transforming industry.