15 Years in 28 Minutes: GPUs Speed Up Simulation of Oil Reservoirs

Italian energy company Eni and U.S.-based Stone Ridge Technology have decreased the time it takes to model an oil reservoir prospect.

Sizing up an oil reservoir just got a lot faster.

Eni, a multinational oil and gas company, holds bragging rights to HPC4, the world’s most powerful commercial supercomputer.

Running 3,200 NVIDIA Tesla GPUs, Eni’s parallel processing beast has completed a breakthrough calculation that highlights an acceleration in reservoir assessment.

Eni’s super computing feat was achieved in partnership with Baltimore-based Stone Ridge Technology, which develops and licenses ECHELON. ECHELON is high-performance NVIDIA GPU-based petroleum reservoir simulation software.

The 18.6 petaflops supercomputer processed 100,000 reservoir models in about 15 and a half hours,  a task which would take 10 days using legacy hardware and software. Each individual model simulated 15 years of production in an average of 28 minutes.

Modeling oil reservoirs is no small computing problem. First exploration experts need to find the reserves by essentially drumming the Earth’s surface and capturing the reflected sound waves.

After massive numerical processing this reflected wave data is turned into images that geoscientists can use to determine if a reservoir prospect contains hydrocarbons and where the hydrocarbons are located within the image. Over the past decade, GPUs have played an increasingly important role in reservoir simulations.

Minimizing Financial, Environmental Risk

Much is at stake for oil companies seeking to responsibly tap new reservoirs or reassess production stage fields at the least financial and environmental risk. Drilling can cost hundreds of millions of dollars. After locating hydrocarbons, quickly determining the most profitable strategies for new or ongoing production matters. For this task oil companies use reservoir simulators.

These reservoir simulators are technical software applications that model how hydrocarbons and water flow under the ground in the presence of wells. They let oil companies such as Eni evaluate virtual production strategies and “what-if” scenarios on supercomputers before committing to a new project as well as readjusting their models on production wells in the real world.

Simulators traditionally run on CPU-based hardware and are limited in both  performance and in the size of the models. It’s not uncommon to have models that take days to run. To improve the odds of hitting production targets, the energy giants are increasingly turning to higher-resolution models and faster software powered by NVIDIA GPUs.

“Stone Ridge was an early adopter of NVIDIA GPUs for high-performance computing and we built ECHELON from the ground up to exploit the technology.” said Vincent Natoli, founder and president of Stone Ridge.  “We see the benefit of that choice now as ECHELON has become faster and more capable with each generation of NVIDIA product. We significantly outpace our CPU-based competitors in both performance and scalability and each year the gap has become more pronounced.”

Simulate Reservoir’s Potential

Eni’s announcement  that its HPC4 supercomputer running ECHELON can speed reservoir simulations highlights new records for the industry. Eni was able to take a high-resolution model of a deep-water reservoir with 5.7 million active cells and generate 100,000 models with varying petro-physical properties. All 100,000 models were completed in 15 hours and 25 minutes, running on Eni’s HPC4 sporting 3,200 Tesla P100 GPUs.

Let’s block ads! (Why?)

The Official NVIDIA Blog

Leave a Reply

%d bloggers like this: