NVIDIA Accelerates Eni’s Oil and Gas Exploration

by Guy Gueritz

Eni has expanded the computing capacity of their Green Data Center with their latest GPU hybrid system, supplied by HPE, called HPC4. It’s packed with 3,200 Tesla GPU accelerators and is set to revolutionize how exploration oil and gas activities are managed.

What Lies Beneath

Finding and producing hydrocarbons is a challenging and difficult process.

A number of techniques have been established to support exploration for new hydrocarbon reserves. The primary one is seismic imaging of the subsurface. Over the last 15 years, leveraging its own internal R&D resources, Eni has developed a number of advanced and fully integrated applications for geological and geophysical studies. One of the most computationally demanding applications in the field of seismic imaging is a proprietary implementation of anisotropic reverse time migration (RTM).

Based on measurements recorded at the surface using an array of receivers, the RTM algorithm creates a very accurate image of the subsurface of the Earth. This is done by modeling how waves travel “forward” through rock layers. The computation is carried out on 3D cubic grids extending over hundreds of square kilometers, down to 10 to 15 kilometers in depth. The outcome is a 3D image of the subsurface of the same size, with typical resolution on the order of 10 to 25 meters, containing several billions of pixels.

This image can then be used by geologists and geophysicists to look for hydrocarbons. Although the name reverse time migration refers to time domain, the results of the method are depth values that can be compared to hard data on rock layers shown by drilling for core samples.

Anisotropic RTM is a particularly useful technology for exploring what lies beneath the Earth’s surface, especially where the geology is truly complex. Such as in areas where subsurface features have steep dips, or in areas with salt tectonic where seismic waves tend to be heavily scattered. In these complex geological settings, the results provided by methods less accurate (and less computationally demanding) than anisotropic RTM are less reliable and hydrocarbon prospects under these features may be obscured or missed at all.

Until recently, the use of anisotropic RTM has been limited to deep-water exploration and subsalt environments, and used to process low-frequency seismic data.

This limitation, imposed due to the computational expense and timing of running RTM on traditional compute clusters, has a severe impact on image resolution and the accuracy of results.

However, using GPU-accelerated computing, Eni has been able to implement the anisotropic RTM process 4-5 times faster than before and can now routinely use this advanced imaging technique in a wide variety of complex geological contexts, at higher frequencies.

Being able to perform anisotropc RTM at higher frequencies is crucial. Although it results in an exponential growth in computational complexity, it often provides more clarity, accuracy and detail. The combination of fast, high-resolution seismic imaging and geological data into a single HPC platform is a key for Eni to produce robust geological models used to explore for new hydrocarbon resources and for reservoir management.

The World’s Most Powerful Industrial System

Now, Eni is firing up their HPC4 supercomputer, which is based at Eni’s Green Data Center in Ferrera Erbognone, just outside Milan. The launch of HPC4 quadruples the company’s computational power and makes its HPC infrastructure the world’s most powerful industrial computing system today.

With 3,200 NVIDIA Tesla P100 GPU accelerators inside, in addition to the existing systems using NVIDIA Tesla K80 GPUs, Eni’s computational peak capacity is now 22.4 petaflops. Alongside the HPC3 supercomputer, the HPC4 will support Eni’s process of digital transformation, accelerating the exploration and development phase of oil and gas reservoirs, and improving the management of the big data generated by operational processes.

“We started our path of supercomputing using hybrid clusters in 2013. We were the first in the O&G industry, and in 2014 we were awarded the HPCwire prize for the best use of HPC in the oil and gas sector,” said Luca Bertelli, chief exploration officer at Eni. “All our HPC systems from the beginning till now have been engineered with the same philosophy of hybrid architecture: powerful and energy-efficient machines that boost the performance of the in-house developed advanced seismic imaging algorithms.

NVIDIA’s Tesla GPU-accelerated computing platforms have been instrumental in supporting Eni’s exploration activity, improving our ability to turn around advanced seismic imaging tasks in a shorter time and with a higher accuracy. GPU-accelerated computing is a key competitive advantage for Eni, enabling us to accelerate the time to market of our discoveries and to shorten our overall upstream cycle.”

Similar Stories